PVF REFERENCE GUIDE
Version 2017
PVF Reference Guide Version 2017 | ii
TABLE OF CONTENTS
Preface............................................................................................................. xiAudience Description.......................................................................................... xiCompatibility and Conformance to Standards............................................................ xiOrganization....................................................................................................xiiHardware and Software Constraints...................................................................... xiiiConventions....................................................................................................xiiiTerms............................................................................................................xivRelated Publications.......................................................................................... xv
Chapter 1. Fortran Data Types................................................................................ 11.1. Fortran Data Types....................................................................................... 1
1.1.1. Fortran Scalars.......................................................................................11.1.2. FORTRAN 77 Aggregate Data Type Extensions.................................................. 31.1.3. Fortran 90 Aggregate Data Types (Derived Types)............................................. 4
Chapter 2. Command-Line Options Reference.............................................................52.1. PGI Compiler Option Summary......................................................................... 5
2.1.1. Build-Related PGI Options......................................................................... 62.1.2. PGI Debug-Related Compiler Options............................................................ 72.1.3. PGI Optimization-Related Compiler Options.................................................... 82.1.4. PGI Linking and Runtime-Related Compiler Options...........................................9
2.2. Generic PGI Compiler Options..........................................................................92.2.1. -#....................................................................................................... 92.2.2. -###...................................................................................................102.2.3. -acc...................................................................................................102.2.4. -Bdynamic........................................................................................... 112.2.5. -Bstatic...............................................................................................122.2.6. -Bstatic_pgi..........................................................................................132.2.7. -byteswapio......................................................................................... 142.2.8. -C..................................................................................................... 142.2.9. -c......................................................................................................152.2.10. -D....................................................................................................162.2.11. -dryrun..............................................................................................172.2.12. -drystdinc.......................................................................................... 172.2.13. -E.................................................................................................... 182.2.14. -F.................................................................................................... 182.2.15. -fast................................................................................................. 192.2.16. -fastsse............................................................................................. 192.2.17. --flagcheck......................................................................................... 202.2.18. -flags................................................................................................202.2.19. -g.................................................................................................... 212.2.20. -gopt................................................................................................ 21
PVF Reference Guide Version 2017 | iii
2.2.21. -help................................................................................................ 222.2.22. -I.....................................................................................................242.2.23. -i2, -i4, -i8.........................................................................................252.2.24. -K<flag>............................................................................................ 262.2.25. --keeplnk........................................................................................... 272.2.26. -L.................................................................................................... 282.2.27. -l<library>.......................................................................................... 282.2.28. -M....................................................................................................292.2.29. -m................................................................................................... 292.2.30. -m64................................................................................................ 302.2.31. -M<pgflag>......................................................................................... 302.2.32. -module <moduledir>............................................................................ 352.2.33. -mp..................................................................................................362.2.34. -noswitcherror.....................................................................................372.2.35. -O<level>........................................................................................... 382.2.36. -o.................................................................................................... 402.2.37. -pc...................................................................................................402.2.38. --pedantic.......................................................................................... 432.2.39. -pgc++libs.......................................................................................... 432.2.40. -pgf77libs...........................................................................................432.2.41. -pgf90libs...........................................................................................442.2.42. -r4 and -r8.........................................................................................442.2.43. -rc................................................................................................... 452.2.44. -S.................................................................................................... 452.2.45. -show............................................................................................... 462.2.46. -silent...............................................................................................462.2.47. -stack............................................................................................... 472.2.48. -ta=tesla(tesla_suboptions),host............................................................... 482.2.49. -time................................................................................................ 512.2.50. -tp <target>[,target...].......................................................................... 512.2.51. -[no]traceback.....................................................................................542.2.52. -u.................................................................................................... 542.2.53. -U....................................................................................................552.2.54. -V[release_number].............................................................................. 562.2.55. -v.................................................................................................... 562.2.56. -W................................................................................................... 572.2.57. -w................................................................................................... 58
2.3. -M Options by Category................................................................................ 582.3.1. Code Generation Controls........................................................................ 582.3.2. Environment Controls............................................................................. 622.3.3. Fortran Language Controls....................................................................... 632.3.4. Inlining Controls....................................................................................672.3.5. Optimization Controls............................................................................. 69
PVF Reference Guide Version 2017 | iv
2.3.6. Miscellaneous Controls............................................................................ 76Chapter 3. Directives Reference............................................................................ 83
3.1. PGI Proprietary Fortran Directive Summary........................................................ 833.1.1. altcode (noaltcode)................................................................................843.1.2. assoc (noassoc)..................................................................................... 853.1.3. bounds (nobounds).................................................................................853.1.4. cncall (nocncall)................................................................................... 853.1.5. concur (noconcur)..................................................................................853.1.6. depchk (nodepchk)................................................................................ 863.1.7. eqvchk (noeqvchk).................................................................................863.1.8. invarif (noinvarif).................................................................................. 863.1.9. ivdep................................................................................................. 863.1.10. lstval (nolstval)................................................................................... 863.1.11. opt.................................................................................................. 863.1.12. prefetch............................................................................................ 873.1.13. safe_lastval........................................................................................ 873.1.14. tp.................................................................................................... 883.1.15. unroll (nounroll).................................................................................. 893.1.16. vector (novector)................................................................................. 893.1.17. vintr (novintr)..................................................................................... 89
3.2. Prefetch Directives and Pragmas..................................................................... 893.3. IGNORE_TKR Directive..................................................................................90
3.3.1. IGNORE_TKR Directive Syntax................................................................... 903.3.2. IGNORE_TKR Directive Format Requirements................................................. 903.3.3. Sample Usage of IGNORE_TKR Directive....................................................... 91
3.4. !DEC\$ Directives........................................................................................ 913.4.1. ALIAS Directive..................................................................................... 913.4.2. ATTRIBUTES Directive............................................................................. 923.4.3. DECORATE Directive............................................................................... 923.4.4. DISTRIBUTE Directive..............................................................................93
Chapter 4. Runtime Environment........................................................................... 944.1. Win64 Programming Model.............................................................................94
4.1.1. Function Calling Sequence....................................................................... 944.1.2. Function Return Values........................................................................... 974.1.3. Argument Passing.................................................................................. 984.1.4. Win64 Fortran Supplement......................................................................100
Chapter 5. PVF Properties.................................................................................. 1065.1. General Property Page................................................................................ 106
5.1.1. General............................................................................................. 1065.1.2. Output Directory..................................................................................1075.1.3. Intermediate Directory.......................................................................... 1075.1.4. Extensions to Delete on Clean................................................................. 1075.1.5. Configuration Type............................................................................... 107
PVF Reference Guide Version 2017 | v
5.1.6. Build Log File..................................................................................... 1075.1.7. Build Log Level................................................................................... 107
5.2. Debugging Property Page............................................................................. 1075.2.1. Debugging.......................................................................................... 1075.2.2. Application Command............................................................................1075.2.3. Application Arguments...........................................................................1085.2.4. Environment....................................................................................... 1085.2.5. Merge Environment...............................................................................1085.2.6. Accelerator Profiling............................................................................. 1085.2.7. MPI Debugging.....................................................................................1095.2.8. Working Directory................................................................................ 1095.2.9. Number of Processes.............................................................................1095.2.10. Working Directory............................................................................... 1095.2.11. Additional Arguments: mpiexec.............................................................. 1105.2.12. Location of mpiexec............................................................................110
5.3. Fortran Property Pages................................................................................1105.4. Fortran | General......................................................................................110
5.4.1. Display Startup Banner.......................................................................... 1105.4.2. Additional Include Directories..................................................................1115.4.3. Module Path....................................................................................... 1115.4.4. Object File Name.................................................................................1115.4.5. Debug Information Format...................................................................... 1125.4.6. Optimization.......................................................................................112
5.5. Fortran | Optimization............................................................................... 1125.5.1. Optimization.......................................................................................1125.5.2. Global Optimizations.............................................................................1135.5.3. Vectorization...................................................................................... 1135.5.4. Inlining..............................................................................................1135.5.5. Use Frame Pointer................................................................................1135.5.6. Loop Unroll Count................................................................................ 1145.5.7. Auto-Parallelization.............................................................................. 114
5.6. Fortran | Preprocessing...............................................................................1145.6.1. Preprocess Source File...........................................................................1145.6.2. Additional Include Directories..................................................................1145.6.3. Ignore Standard Include Path...................................................................1155.6.4. Preprocessor Definitions.........................................................................1155.6.5. Undefine Preprocessor Definitions............................................................. 115
5.7. Fortran | Code Generation...........................................................................1165.7.1. Runtime Library...................................................................................116
5.8. Fortran | Language....................................................................................1165.8.1. Fortran Dialect....................................................................................1165.8.2. Treat Backslash as Character................................................................... 1165.8.3. Extend Line Length...............................................................................117
PVF Reference Guide Version 2017 | vi
5.8.4. Enable OpenMP Directives...................................................................... 1175.8.5. Enable OpenACC Directives..................................................................... 1175.8.6. OpenACC Autoparallelization................................................................... 1175.8.7. OpenACC Required............................................................................... 1185.8.8. OpenACC Routineseq............................................................................. 1185.8.9. OpenACC Wait.....................................................................................1185.8.10. OpenACC Conformance Level................................................................. 1185.8.11. OpenACC Sync................................................................................... 1195.8.12. MPI.................................................................................................1195.8.13. Enable CUDA Fortran........................................................................... 1195.8.14. CUDA Fortran Register Limit.................................................................. 1205.8.15. CUDA Fortran Use Fused Multiply-Adds......................................................1205.8.16. CUDA Fortran Use Fast Math Library........................................................ 1205.8.17. CUDA Fortran Debug............................................................................1205.8.18. CUDA Fortran Line Information............................................................... 1215.8.19. CUDA Fortran Use LLVM Back End............................................................ 1215.8.20. CUDA Fortran Unroll............................................................................ 1215.8.21. CUDA Fortran Flush to Zero...................................................................1215.8.22. CUDA Fortran Toolkit........................................................................... 1225.8.23. CUDA Fortran Compute Capability........................................................... 1225.8.24. CUDA Fortran Fermi............................................................................ 1235.8.25. CUDA Fortran Fermi+...........................................................................1235.8.26. CUDA Fortran Kepler........................................................................... 1235.8.27. CUDA Fortran Kepler+.......................................................................... 1235.8.28. CUDA Fortran Keep Binary.....................................................................1235.8.29. CUDA Fortran Keep Kernel Source........................................................... 1235.8.30. CUDA Fortran Keep PTX........................................................................1245.8.31. CUDA Fortran Keep PTXAS.....................................................................1245.8.32. CUDA Fortran Generate RDC.................................................................. 1245.8.33. CUDA Fortran Emulation....................................................................... 1245.8.34. CUDA Fortran Madconst........................................................................ 124
5.9. Fortran | Floating Point Options.................................................................... 1245.9.1. Floating Point Exception Handling.............................................................1255.9.2. Floating Point Consistency...................................................................... 1255.9.3. Flush Denormalized Results to Zero...........................................................1255.9.4. Treat Denormalized Values as Zero............................................................1255.9.5. IEEE Arithmetic................................................................................... 125
5.10. Fortran | External Procedures..................................................................... 1265.10.1. Calling Convention.............................................................................. 1265.10.2. String Length Arguments.......................................................................1265.10.3. Case of External Names........................................................................126
5.11. Fortran | Libraries................................................................................... 1275.11.1. Use MKL...........................................................................................127
PVF Reference Guide Version 2017 | vii
5.12. Fortran | Target Processors.........................................................................1275.12.1. AMD Athlon.......................................................................................1275.12.2. AMD Barcelona...................................................................................1275.12.3. AMD Bulldozer................................................................................... 1285.12.4. AMD Istanbul..................................................................................... 1285.12.5. AMD Piledriver................................................................................... 1285.12.6. AMD Shanghai.................................................................................... 1285.12.7. Intel Core 2...................................................................................... 1285.12.8. Intel Core i7..................................................................................... 1285.12.9. Intel Penryn...................................................................................... 1285.12.10. Intel Pentium 4................................................................................ 1285.12.11. Intel Sandy Bridge............................................................................. 1285.12.12. Generic x86-64 [x64 only]....................................................................129
5.13. Fortran | Target Accelerators...................................................................... 1295.13.1. Target NVIDIA Tesla............................................................................. 1295.13.2. Tesla Register Limit.............................................................................1295.13.3. Tesla Use Fused Multiply-Adds................................................................ 1305.13.4. Tesla Use Fast Math Library................................................................... 1305.13.5. Tesla LLVM........................................................................................ 1305.13.6. Tesla Noattach...................................................................................1305.13.7. Tesla Pin Host Memory......................................................................... 1305.13.8. Tesla Autocollapse.............................................................................. 1315.13.9. Tesla Debug...................................................................................... 1315.13.10. Tesla Lineinfo...................................................................................1315.13.11. Tesla Unroll..................................................................................... 1315.13.12. Tesla Required..................................................................................1325.13.13. Tesla Flush to Zero............................................................................ 1325.13.14. Tesla Generate RDC........................................................................... 1325.13.15. Tesla CUDA Toolkit.............................................................................1325.13.16. Tesla Compute Capability.................................................................... 1335.13.17. Tesla CC Fermi................................................................................. 1335.13.18. Tesla CC Fermi+................................................................................1335.13.19. Tesla CC Kepler................................................................................ 1345.13.20. Tesla CC Kepler+............................................................................... 1345.13.21. Tesla: Keep Kernel Files...................................................................... 134
5.14. Fortran | Diagnostics................................................................................ 1345.14.1. Warning Level....................................................................................1345.14.2. Generate Assembly............................................................................. 1345.14.3. Annotate Assembly..............................................................................1355.14.4. Accelerator Information........................................................................1355.14.5. CCFF Information................................................................................1355.14.6. Fortran Language Information................................................................ 1355.14.7. Inlining Information.............................................................................135
PVF Reference Guide Version 2017 | viii
5.14.8. IPA Information.................................................................................. 1355.14.9. Loop Intensity Information.................................................................... 1355.14.10. Loop Optimization Information..............................................................1355.14.11. LRE Information................................................................................ 1365.14.12. OpenMP Information...........................................................................1365.14.13. Optimization Information.....................................................................1365.14.14. Parallelization Information................................................................... 1365.14.15. Unified Binary Information................................................................... 1365.14.16. Vectorization Information.................................................................... 136
5.15. Fortran | Profiling....................................................................................1365.15.1. Suppress CCFF Information.................................................................... 1365.15.2. Enable Limited DWARF......................................................................... 137
5.16. Fortran | Runtime....................................................................................1375.16.1. Check Array Bounds............................................................................ 1375.16.2. Check Pointers...................................................................................1375.16.3. Check Stack...................................................................................... 1375.16.4. Command Line...................................................................................137
5.17. Fortran | Command Line............................................................................1385.17.1. Command Line...................................................................................138
5.18. Linker Property Pages................................................................................1385.19. Linker | General......................................................................................138
5.19.1. Output File....................................................................................... 1385.19.2. Additional Library Directories.................................................................1395.19.3. Stack Reserve Size.............................................................................. 1395.19.4. Stack Commit Size.............................................................................. 1395.19.5. Export Symbols.................................................................................. 139
5.20. Linker | Input.........................................................................................1395.20.1. Additional Dependencies.......................................................................139
5.21. Linker | Command Line............................................................................. 1405.21.1. Command Line...................................................................................140
5.22. Librarian Property Pages............................................................................ 1405.23. Librarian | General.................................................................................. 140
5.23.1. Output File....................................................................................... 1405.23.2. Additional Library Directories.................................................................1415.23.3. Additional Dependencies.......................................................................141
5.24. Librarian | Command Line..........................................................................1425.24.1. Command Line...................................................................................142
5.25. Resources Property Page............................................................................ 1425.26. Resources | Command Line.........................................................................142
5.26.1. Command Line...................................................................................1425.27. Build Events Property Page......................................................................... 142
5.27.1. Build Event....................................................................................... 1435.27.2. Command Line...................................................................................143
PVF Reference Guide Version 2017 | ix
5.27.3. Description....................................................................................... 1435.27.4. Excluded From Build............................................................................143
5.28. Custom Build Step Property Page..................................................................1435.28.1. Custom Build Step | General................................................................. 1435.28.2. Command Line...................................................................................1435.28.3. Description....................................................................................... 1445.28.4. Outputs........................................................................................... 1445.28.5. Additional Dependencies.......................................................................144
Chapter 6. PVF Build Macros............................................................................... 145Chapter 7. Fortran Module/Library Interfaces for Windows..........................................148
7.1. Source Files............................................................................................. 1487.2. Data Types.............................................................................................. 1487.3. Using DFLIB, LIBM, and DFPORT.....................................................................149
7.3.1. DFLIB................................................................................................1497.3.2. LIBM................................................................................................. 1507.3.3. DFPORT............................................................................................. 151
7.4. Using the DFWIN module............................................................................. 1567.5. Supported Libraries and Modules....................................................................156
7.5.1. advapi32............................................................................................1577.5.2. comdlg32........................................................................................... 1587.5.3. dfwbase............................................................................................ 1597.5.4. dfwinty............................................................................................. 1597.5.5. gdi32................................................................................................ 1597.5.6. kernel32............................................................................................ 1627.5.7. shell32.............................................................................................. 1697.5.8. user32...............................................................................................1697.5.9. winver.............................................................................................. 1747.5.10. wsock32........................................................................................... 174
Chapter 8. Messages..........................................................................................1758.1. Diagnostic Messages................................................................................... 1758.2. Phase Invocation Messages........................................................................... 1768.3. Fortran Compiler Error Messages....................................................................176
8.3.1. Message Format...................................................................................1768.3.2. Message List....................................................................................... 176
8.4. Fortran Run-time Error Messages....................................................................2128.4.1. Message Format...................................................................................2128.4.2. Message List....................................................................................... 212
Chapter 9. Contact Information............................................................................215
PVF Reference Guide Version 2017 | x
LIST OF TABLES
Table 1 PGI Compilers and Commands ..................................................................... xiv
Table 2 Representation of Fortran Data Types .............................................................. 1
Table 3 Real Data Type Ranges ................................................................................ 2
Table 4 Scalar Type Alignment .................................................................................2
Table 5 PGI Build-Related Compiler Options ................................................................ 6
Table 6 PGI Debug-Related Compiler Options ............................................................... 8
Table 7 Optimization-Related PGI Compiler Options .......................................................8
Table 8 Linking and Runtime-Related PGI Compiler Options ..............................................9
Table 9 Subgroups for -help Option ......................................................................... 23
Table 10 -M Options Summary ................................................................................30
Table 11 Optimization and -O, -g, -Mvect, and -Mconcur Options ..................................... 39
Table 12 IGNORE_TKR Example ...............................................................................91
Table 13 Register Allocation .................................................................................. 95
Table 14 Standard Stack Frame .............................................................................. 95
Table 15 Register Allocation for Example A-4 ............................................................. 99
Table 16 Win64 Fortran Fundamental Types ..............................................................101
Table 17 Fortran and C/C++ Data Type Compatibility .................................................. 103
Table 18 Fortran and C/C++ Representation of the COMPLEX Type ...................................103
Table 19 PVF Build Macros ...................................................................................145
Table 20 Fortran Data Type Mappings ..................................................................... 148
Table 21 DFLIB Function Summary ......................................................................... 149
Table 22 LIBM Functions ......................................................................................150
Table 23 DFPORT Functions ..................................................................................151
PVF Reference Guide Version 2017 | xi
PREFACE
This guide is part of a set of manuals that describe how to use the PGI Fortran compilersand program development tools integrated with Microsoft Visual Studio. These tools,combined with Visual Studio and assorted libraries, are collectively known as PGI VisualFortran®, or PVF®. You can use PVF to edit, compile, debug, optimize, and profile serialand parallel applications for x64 processor-based systems.
The PGI Visual Fortran Reference Manual is the reference companion to the PGI VisualFortran User’s Guide which provides operating instructions for both the Visual Studiointegrated development environment as well as command-level compilation and generalinformation about PGI’s compilers. Neither guide teaches the Fortran programminglanguage.
Audience DescriptionThis manual is intended for scientists and engineers using PGI Visual Fortran. To fullyunderstand this guide, you should be aware of the role of high-level languages, suchas Fortran, in the software development process; and you should have some level ofunderstanding of programming. PGI Visual Fortran is available on a variety of x86-64/x64 hardware platforms and variants of the Windows operating system. You need to befamiliar with the basic commands available on your system.
Compatibility and Conformance to StandardsYour system needs to be running a properly installed and configured version of this PGIproduct. For information on installing PVF, refer to the Release Notes and InstallationGuide included with your software.
For further information, refer to the following:
‣ American National Standard Programming Language FORTRAN, ANSI X3. -1978 (1978).‣ ISO/IEC 1539-1 : 1991, Information technology – Programming Languages – Fortran,
Geneva, 1991 (Fortran 90).‣ ISO/IEC 1539-1 : 1997, Information technology – Programming Languages – Fortran,
Geneva, 1997 (Fortran 95).
Preface
PVF Reference Guide Version 2017 | xii
‣ ISO/IEC 1539-1 : 2004, Information technology – Programming Languages – Fortran,Geneva, 2004 (Fortran 2003).
‣ ISO/IEC 1539-1 : 2010, Information technology – Programming Languages – Fortran,Geneva, 2010 (Fortran 2008).
‣ Fortran 95 Handbook Complete ISO/ANSI Reference, Adams et al, The MIT Press,Cambridge, Mass, 1997.
‣ The Fortran 2003 Handbook, Adams et al, Springer, 2009.‣ OpenMP Application Program Interface, Version 3.1, July 2011, http://
www.openmp.org.‣ Programming in VAX Fortran, Version 4.0, Digital Equipment Corporation
(September, 1984).‣ IBM VS Fortran, IBM Corporation, Rev. GC26-4119.‣ Military Standard, Fortran, DOD Supplement to American National Standard
Programming Language Fortran, ANSI x.3-1978, MIL-STD-1753 (November 9, 1978).‣ ISO/IEC 9899:2011, Information Technology – Programming Languages – C, Geneva,
2011 (C11).‣ ISO/IEC 14882:2011, Information Technology – Programming Languages – C++,
Geneva, 2011 (C++11).
OrganizationUsers typically begin by wanting to know how to use a product and often then find thatthey need more information and facts about specific areas of the product. Knowing howas well as why you might use certain options or perform certain tasks is key to usingthe PGI compilers and tools effectively and efficiently. However, once you have thisknowledge and understanding, you very likely might find yourself wanting to knowmuch more about specific areas or specific topics.
To facilitate ease of use, this manual contains detailed reference information aboutspecific aspects of the compiler, such as the details of compiler options, directives, andmore. This guide contains these sections:
Fortran Data Types describes the data types that are supported by the PGI Fortrancompilers.
Command-Line Options Reference provides a detailed description of each command-line option.
Directives Reference contains detailed descriptions of PGI’s proprietary directives.
Runtime Environment describes the programming model supported for compiler codegeneration, including register conventions and calling conventions for x64 processor-based systems running a Windows operating system.
PVF Properties provides a description of Property Pages that PGI supports.
PVF Build Macros provides a description of the build macros that PVF supports.
Fortran Module/Library Interfaces for Windows provides a description of the Fortranmodule library interfaces that PVF supports.
Preface
PVF Reference Guide Version 2017 | xiii
Messages provides a list of compiler error messages.
Hardware and Software ConstraintsThis guide describes versions of the PGI Visual Fortran that are intended for use on x64processor-based systems. Details concerning environment-specific values and defaultsand system-specific features or limitations are presented in the release notes deliveredwith the PGI Visual Fortran.
ConventionsThis guide uses the following conventions:italic
is used for emphasis.Constant Width
is used for filenames, directories, arguments, options, examples, and for languagestatements in the text, including assembly language statements.
Boldis used for commands.
[ item1 ]in general, square brackets indicate optional items. In this case item1 is optional. Inthe context of p/t-sets, square brackets are required to specify a p/t-set.
{ item2 | item 3 }braces indicate that a selection is required. In this case, you must select either item2 oritem3.
filename ...ellipsis indicate a repetition. Zero or more of the preceding item may occur. In thisexample, multiple filenames are allowed.
FORTRANFortran language statements are shown in the text of this guide using a reduced fixedpoint size.
C/C++C/C++ language statements are shown in the test of this guide using a reduced fixedpoint size.
The PGI compilers and tools are supported on a wide variety of Linux, macOS andWindows operating systems running on 64-bit x86-compatible processors, and on Linuxrunning on OpenPOWER processors. (Currently, the PGI debugger is supported onx86-64/x64 only.) See the Compatibility and Installation section on the PGI website athttps://www.pgroup.com/products/index.htm?tab=compat for a comprehensive listingof supported platforms.
Support for 32-bit development was deprecated in PGI 2016 and is no longer availableas of the PGI 2017 release. PGI 2017 is only available for 64-bit operating systems anddoes not include the ability to compile 32-bit applications for execution on either 32-or 64-bit operating systems.
Preface
PVF Reference Guide Version 2017 | xiv
TermsA number of terms related to systems, processors, compilers and tools are usedthroughout this guide. For example:
accelerator FMA -mcmodel=medium static linking
AVX host -mcmodel=small Win32
CUDA hyperthreading (HT) MPI Win64
device large arrays multicore x64
DLL license keys NUMA s86
driver LLVM SIMD x87
DWARF manycore SSE
For a complete definition of these terms and other terms in this guide with which youmay be unfamiliar, please refer to the PGI online glossary at https://www.pgroup.com/support/definitions.htm.
The following table lists the PGI compilers and tools and their correspondingcommands:
Table 1 PGI Compilers and Commands
Compiler or Tool Language or Function Command
PGF77 ANSI FORTRAN 77 pgf77
PGFORTRAN ISO/ANSI Fortran 2003 pgfortran
PGI Debugger Source code debugger pgdbg
PGI Profiler Performance profiler pgprof
In general, the designation PGI Fortran is used to refer to the PGI Fortran 2003 compiler,and pgfortran is used to refer to the command that invokes the compiler. A similarconvention is used for each of the PGI compilers and tools.
For simplicity, examples of command-line invocation of the compilers generallyreference the pgfortran command, and most source code examples are written inFortran. Usage of the PGF77 compiler, whose features are a subset of PGFORTRAN, issimilar.
There are a wide variety of 64-bit x86-compatible processors in use. All are supported bythe PGI compilers and tools. Most of these processors are forward-compatible, but notbackward-compatible, meaning that code compiled to target a given processor will notnecessarily execute correctly on a previous-generation processor.
A table listing the processor options that PGI supports is available in the Release Notes.The table also includes the features utilized by the PGI compilers that distinguish themfrom a compatibility standpoint.
Preface
PVF Reference Guide Version 2017 | xv
In this manual, the convention is to use "x86" to specify the group of processors thatare "32-bit" but not "64-bit". The convention is to use "x64" to specify the group ofprocessors that are both "32-bit" and "64-bit". x86 processor-based systems can run only32-bit operating systems. x64 processor-based systems can run either 32-bit or 64-bitoperating systems, and can execute all 32-bit x86 binaries in either case. x64 processorshave additional registers and 64-bit addressing capabilities that are utilized by the PGIcompilers and tools when running on a 64-bit operating system. The prefetch, SSE1,SSE2, SSE3, and AVX processor features further distinguish the various processors.Where such distinctions are important with respect to a given compiler option orfeature, it is explicitly noted in this manual.
The default for performing scalar floating-point arithmetic is to use SSE instructionson targets that support SSE1 and SSE2.
Support for 32-bit development was deprecated in PGI 2016 and is no longer availableas of the PGI 2017 release. PGI 2017 is only available for 64-bit operating systems anddoes not include the ability to compile 32-bit applications for execution on either 32-bit or 64-bit operating systems.
Related PublicationsThe following documents contain additional information related to the x86-64 and x64architectures, and the compilers and tools available from The Portland Group.
‣ PGI Fortran Reference Manual, https://www.pgroup.com/resources/docs.phpdescribes the FORTRAN 77, Fortran 90/95, Fortran 2003 statements, data types,input/output format specifiers, and additional reference material related to use ofthe PGI Fortran compilers.
‣ System V Application Binary Interface Processor Supplement by AT&T UNIX SystemLaboratories, Inc. (Prentice Hall, Inc.).
‣ System V Application Binary Interface X86-64 Architecture Processor Supplement, http://www.x86-64.org/documentation_folder/abi.pdf.
‣ Fortran 95 Handbook Complete ISO/ANSI Reference, Adams et al, The MIT Press,Cambridge, Mass, 1997.
‣ Programming in VAX Fortran, Version 4.0, Digital Equipment Corporation (September,1984).
‣ IBM VS Fortran, IBM Corporation, Rev. GC26-4119.
Preface
PVF Reference Guide Version 2017 | xvi
PVF Reference Guide Version 2017 | 1
Chapter 1.FORTRAN DATA TYPES
This section describes the scalar and aggregate data types recognized by the PGI Fortrancompilers, the format and alignment of each type in memory, and the range of valueseach type can have on 64-bit operating systems.
1.1. Fortran Data Types
1.1.1. Fortran ScalarsA scalar data type holds a single value, such as the integer value 42 or the real value112.6. The next table lists scalar data types, their size, format and range. Table 3 showsthe range and approximate precision for Fortran real data types. Table 4 shows thealignment for different scalar data types. The alignments apply to all scalars, whetherthey are independent or contained in an array, a structure or a union.
Table 2 Representation of Fortran Data Types
Fortran Data Type Format Range
INTEGER 2's complement integer -231 to 231-1
INTEGER*2 2's complement integer -32768 to 32767
INTEGER*4 2's complement integer -231 to 231-1
INTEGER*8 2's complement integer -263 to 263-1
LOGICAL 32-bit value true or false
LOGICAL*1 8-bit value true or false
LOGICAL*2 16-bit value true or false
LOGICAL*4 32-bit value true or false
LOGICAL*8 64-bit value true or false
BYTE 2's complement -128 to 127
REAL Single-precision floating point 10-37 to 1038 (1)
Fortran Data Types
PVF Reference Guide Version 2017 | 2
Fortran Data Type Format Range
REAL*4 Single-precision floating point 10-37 to 10 38 (1)
REAL*8 Double-precision floating point 10-307 to 10 308 (1)
DOUBLE PRECISION Double-precision floating point 10-307 to 10308 (1)
COMPLEX Single-precision floating point 10-37 to 1038 (1)
DOUBLE COMPLEX Double-precision floating point 10-307 to 10308 (1)
COMPLEX*16 Double-precision floating point 10-307 to 10308 (1)
CHARACTER*n Sequence of n bytes
(1) Approximate value
The logical constants .TRUE. and .FALSE. are all ones and all zeroes, respectively.Internally, the value of a logical variable is true if the least significant bit is one and falseotherwise. When the option -Munixlogical is set, a logical variable with a non-zerovalue is true and with a zero value is false.
A variable of logical type may appear in an arithmetic context, and the logical type isthen treated as an integer of the same size.
Table 3 Real Data Type Ranges
Data Type Binary Range Decimal Range Digits of Precision
REAL -2-126 to 2128 10-37 to 1038 (1) 7–8
REAL*8 -2-1022 to 21024 10-307 to 10308 (1) 15–16
Table 4 Scalar Type Alignment
This Type... ...Is aligned on this size boundary
LOGICAL*1 1-byte
LOGICAL*2 2-byte
LOGICAL*4 4-byte
LOGICAL*8 8-byte
BYTE 1-byte
INTEGER*2 2-byte
INTEGER*4 4-byte
INTEGER*8 8-byte
REAL*4 4-byte
REAL*8 8-byte
COMPLEX*8 4-byte
COMPLEX*16 8-byte
Fortran Data Types
PVF Reference Guide Version 2017 | 3
1.1.2. FORTRAN 77 Aggregate Data Type ExtensionsThe PGFORTRAN compiler supports de facto standard extensions to FORTRAN 77 thatallow for aggregate data types. An aggregate data type consists of one or more scalardata type objects. You can declare the following aggregate data types:
‣ An array consists of one or more elements of a single data type placed in contiguouslocations from first to last.
‣ A structure can contain different data types. The members are allocated in the orderthey appear in the definition but may not occupy contiguous locations.
‣ A union is a single location that can contain any of a specified set of scalar oraggregate data types. A union can have only one value at a time. The data type ofthe union member to which data is assigned determines the data type of the unionafter that assignment.
The alignment of an array, a structure or union (an aggregate) affects how much spacethe object occupies and how efficiently the processor can address members. Arrays usethe alignment of their members.Array types
align according to the alignment of the array elements. For example, an array ofINTEGER*2 data aligns on a 2-byte boundary.
Structures and Unionsalign according to the alignment of the most restricted data type of the structureor union. In the next example, the union aligns on a 4-byte boundary since thealignment of c, the most restrictive element, is four.
STRUCTURE /astr/UNION MAP INTEGER*2 a ! 2 bytes END MAP MAP BYTE b ! 1 byte END MAP MAP INTEGER*4 c ! 4 bytes END MAPEND UNIONEND STRUCTURE
Structure alignment can result in unused space called padding. Padding betweenmembers of the structure is called internal padding. Padding between the last memberand the end of the space is called tail padding.
The offset of a structure member from the beginning of the structure is a multiple of themember's alignment. For example, since an INTEGER*2 aligns on a 2-byte boundary, theoffset of an INTEGER*2 member from the beginning of a structure is a multiple of twobytes.
Fortran Data Types
PVF Reference Guide Version 2017 | 4
1.1.3. Fortran 90 Aggregate Data Types (Derived Types)The Fortran 90 standard added formal support for aggregate data types. The TYPEstatement begins a derived type data specification or declares variables of a specifieduser-defined type. For example, the following would define a derived type ATTENDEE:TYPE ATTENDEE CHARACTER(LEN=30) NAME CHARACTER(LEN=30) ORGANIZATION CHARACTER (LEN=30) EMAILEND TYPE ATTENDEE
In order to declare a variable of type ATTENDEE and access the contents of such avariable, code such as the following would be used:TYPE (ATTENDEE) ATTLIST(100). . .ATTLIST(1)%NAME = ‘JOHN DOE’
PVF Reference Guide Version 2017 | 5
Chapter 2.COMMAND-LINE OPTIONS REFERENCE
A command-line option allows you to specify specific behavior when a program iscompiled and linked. Compiler options perform a variety of functions, such as settingcompiler characteristics, describing the object code to be produced, controlling thediagnostic messages emitted, and performing some preprocessor functions. Mostoptions that are not explicitly set take the default settings. This reference sectiondescribes the syntax and operation of each compiler option. For easy reference, theoptions are arranged in alphabetical order.
For an overview and tips on options usage and which options are best for which tasks,refer to the ‘Using Command-line Options’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php, which also provides summary tables of thedifferent options.
This section uses the following notation:[item]
Square brackets indicate that the enclosed item is optional.{item | item}
Braces indicate that you must select one and only one of the enclosed items. A verticalbar (|) separates the choices.
...Horizontal ellipses indicate that zero or more instances of the preceding item arevalid.
2.1. PGI Compiler Option SummaryThe following tables include all the PGI compiler options that are not language-specific.The options are separated by category for easier reference.
For a complete description of each option, refer to the detailed information later in thissection.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 6
2.1.1. Build-Related PGI OptionsThe options included in the following table pertain to the initial building of yourprogram or application.
Table 5 PGI Build-Related Compiler Options
Option Description
-# Display invocation information.
-### Shows but does not execute the driver commands (same as theoption -dryrun).
-acc Enable OpenACC directives.
-Bdynamic Compiles for and links to the shared object version of the PGIruntime libraries.
-Bstatic_pgi Compiles for and links to the static version of the PGI runtimelibraries.
-c Stops after the assembly phase and saves the object code infilename.o.
-D<args> Defines a preprocessor macro.
-dryrun Shows but does not execute driver commands.
-drystdinc Displays the standard include directories and then exits thecompiler.
-E Stops after the preprocessing phase and displays the preprocessedfile on the standard output.
-F Stops after the preprocessing phase and saves the preprocessedfile in filename.f. This option is only valid for the PGI Fortrancompilers.
--flagcheck Simply return zero status if flags are correct.
-flags Display valid driver options.
-g77libs (Linux only) Allow object files generated by g77 to be linked intoPGI main programs.
-I<dirname> Adds a directory to the search path for #include files.
-i2: Treat INTEGER variables as 2 bytes.
-i4: Treat INTEGER variables as 4 bytes.
-i2, -i4 and -i8
-i8: Treat INTEGER and LOGICAL variables as 8 bytes and use 64-bitsfor INTEGER*8 operations.
-K<flag> Requests special compilation semantics with regard to conformanceto IEEE 754.
--keeplnk If the compiler generates a temporary indirect file for a long linkercommand, preserves the temporary file instead of deleting it.
-L<dirname> Specifies a directory to search for libraries.
-l<library> Loads a library.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 7
Option Description
-m Displays a link map on the standard output.
-M<pgflag> Selects variations for code generation and optimization.
-mcmodel=medium (-tp k8-64 and -tp p7-64 targets only) Generate code whichsupports the medium memory model in the linux86-64 environment.
-module <moduledir> Save/search for module files in directory <moduledir>.
-mp[=all, align,bind,[no]numa] Interpret and process user-inserted shared-memory parallelprogramming directives.
-noswitcherror Ignore unknown command line switches after printing an warningmessage.
-o Names the object file.
-pc <val> (-tp px/p5/p6/piii targets only) Set precision globally for x87floating-point calculations; must be used when compiling the mainprogram. <val> may be one of 32, 64 or 80.
-pgf77libs Append PGF77 runtime libraries to the link line.
-pgf90libs Append PGF90/PGF95/PGFORTRAN runtime libraries to the link line.
-r4: Interpret DOUBLE PRECISION variables as REAL.-r4 and -r8
-r8: Interpret REAL variables as DOUBLE PRECISION.
-rc file Specifies the name of the driver's startup file.
-S Stops after the compiling phase and saves the assembly-languagecode in filename.s.
-show Display driver's configuration parameters after startup.
-silent Do not print warning messages.
-time Print execution times for the various compilation steps.
-tp <target> [,target...] Specify the type(s) of the target processor(s).
-u<symbol> Initializes the symbol table with <symbol>, which is undefined forthe linker. An undefined symbol triggers loading of the first memberof an archive library.
-U<symbol> Undefine a preprocessor macro.
-V[release_number] Displays the version messages and other information, or allowsinvocation of a version of the compiler other than the default.
-v Displays the compiler, assembler, and linker phase invocations.
-W Passes arguments to a specific phase.
-w Do not print warning messages.
-Xlinker <option> Passes options to the linker.
2.1.2. PGI Debug-Related Compiler OptionsThe options included in the following table pertain to debugging your program orapplication.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 8
Table 6 PGI Debug-Related Compiler Options
Option Description
-C (Fortran only) Generates code to check array bounds.
-c Instrument the generated executable to perform array boundschecking at runtime.
-E Stops after the preprocessing phase and displays the preprocessedfile on the standard output.
--flagcheck Simply return zero status if flags are correct.
-flags Display valid driver options.
-g Includes debugging information in the object module; sets theoptimization level to zero unless a -O option is present on thecommand line.
-gopt Includes debugging information in the object module, but forcesassembly code generation identical to that obtained when -goptis not present on the command line.
-K<flag> Requests special compilation semantics with regard to conformanceto IEEE 754.
--keeplnk If the compiler generates a temporary indirect file for a long linkercommand, preserves the temporary file instead of deleting it.
-M<pgflag> Selects variations for code generation and optimization.
-pc <val> (-tp px/p5/p6/piii targets only) Set precision globally for x87floating-point calculations; must be used when compiling the mainprogram. <val> may be one of 32, 64 or 80.
-[no]traceback Adds debug information for runtime traceback for use with theenvironment variable PGI_TERM.
2.1.3. PGI Optimization-Related Compiler OptionsThe options included in the following table pertain to optimizing your program orapplication code.
Table 7 Optimization-Related PGI Compiler Options
Option Description
-fast Generally optimal set of flags.
-fastsse Generally optimal set of flags for targets that include SSE/SSE2capability.
-M<pgflag> Selects variations for code generation and optimization.
-mp[=all, align,bind,[no]numa] Interpret and process user-inserted shared-memory parallelprogramming directives.
-O<level> Specifies code optimization level where <level> is 0, 1, 2, 3, or 4.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 9
Option Description
-pc <val> (-tp px/p5/p6/piii targets only) Set precision globally for x87floating-point calculations; must be used when compiling the mainprogram. <val> may be one of 32, 64 or 80.
2.1.4. PGI Linking and Runtime-Related CompilerOptionsThe options included in the following table pertain to defining parameters related tolinking and running your program or application.
Table 8 Linking and Runtime-Related PGI Compiler Options
Option Description
-Bdynamic Compiles for and links to the DLL version of the PGI runtimelibraries.
-Bstatic_pgi Compiles for and links to the static version of the PGI runtimelibraries.
-byteswapio (Fortran only) Swap bytes from big-endian to little-endian or viceversa on input/output of unformatted data.
-i2: Treat INTEGER variables as 2 bytes.
-i4: Treat INTEGER variables as 4 bytes.
-i2, -i4 and -i8
-i8: Treat INTEGER and LOGICAL variables as 8 bytes and use 64-bitsfor INTEGER*8 operations.
-K<flag> Requests special compilation semantics with regard to conformanceto IEEE 754.
-M<pgflag> Selects variations for code generation and optimization.
-mcmodel=medium (-tp k8-64 and -tp p7-64 targets only) Generate code which supportsthe medium memory model in the linux86-64 environment.
-Xlinker <option> Pass options to the linker.
2.2. Generic PGI Compiler OptionsThe following descriptions are for all the PGI options. For easy reference, the optionsare arranged in alphabetical order. For a list of options by tasks, refer to the tables in thebeginning of this section.
2.2.1. -#Displays the invocations of the compiler, assembler and linker.
Default
The compiler does not display individual phase invocations.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 10
Usage
The following command-line requests verbose invocation information.$ pgfortran -# prog.f
Description
The -# option displays the invocations of the compiler, assembler and linker. Theseinvocations are command-lines created by the driver from your command-line input andthe default value.
Related options
-Minfo[=option [,option,...]], -V[release_number], -v
2.2.2. -###Displays the invocations of the compiler, assembler and linker, but does not executethem.
Default
The compiler does not display individual phase invocations.
Usage
The following command-line requests verbose invocation information.$ pgfortran -### myprog.f
Description
Use the -### option to display the invocations of the compiler, assembler and linkerbut not to execute them. These invocations are command lines created by the compilerdriver from the rc files and the command-line options.
Related options
-#, -dryrun, -Minfo[=option [,option,...]], -V[release_number]
2.2.3. -accEnables OpenACC directives.
Default
The compiler enables OpenACC directives.
Syntax-acc[=[no]autopar|[no]required|strict|verystrict]
Command-Line Options Reference
PVF Reference Guide Version 2017 | 11
[no]autoparEnable [default] loop autoparallelization within acc parallel. The default is to autopar,that is, to enable loop autoparallelization.
[no]requiredInstructs the compiler to issue a compiler error if the compute regions fail toaccelerate. The default is required.
strictInstructs the compiler to issue warnings for non-OpenACC accelerator directives.
verystrictInstructs the compiler to fail with an error for any non-OpenACC acceleratordirective.
Usage
The following command-line requests that OpenACC directives be enabled and that anerror be issued for any non-OpenACC accelerator directive.$ pgfortran -acc=verystrict -g prog.f
Description
The -acc option enables OpenACC directives. You can use the suboptions to specify loopautoparallelization, how the compiler reports compute regions failures to accelerate, andwhether to issue a warning or an error for non-OpenACC accelerator directives.
Starting in PGI 14.1, you control the OpenACC compiler behavior related to acceleratorcode generation failures with the required suboption. The OpenACC compilers nowissue a compile-time error if accelerator code generation fails. In previous releases, thecompiler would issue a warning, then generate code to run the compute kernel on thehost. This previous behavior generates incorrect results if the compute kernels are insidea data region and the host and device memory values are inconsistent. You can enablethe old behavior by using the -acc norequired switch.
Related options
-g, -ta=tesla(tesla_suboptions),host
2.2.4. -BdynamicCompiles for and links to the shared object version of the PGI runtime libraries.
Default
The compiler uses static libraries.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 12
Usage
On Windows, you can create the DLL obj1.dll and its import library obj1.lib usingthe following series of commands:% pgfortran -Bdynamic -c object1.f% pgfortran -Mmakedll object1.obj -o obj1.dll
Then compile the main program using this command:$ pgfortran -# prog.f
For a complete example in Windows, refer to the example: ‘Build a DLL: Fortran’ in the‘Creating and Using Libraries’ section of the PGI Compiler User’s Guide.
Description
Use this option to compile for and link to the shared object version of the PGI runtimelibraries. This flag is required when linking with any DLL built by the PGI compilers.For Windows, this flag corresponds to the /MD flag used by Microsoft’s cl compilers.
On Windows, -Bdynamic must be used for both compiling and linking.
When you use the PGI compiler flag -Bdynamic to create an executable that links tothe shared object form of the runtime, the executable built is smaller than one builtwithout -Bdynamic. The PGI runtime shared object(s), however, must be available onthe system where the executable is run. The -Bdynamic flag must be used when anexecutable is linked against a shared object built by the PGI compilers.
Related options
-Bstatic , -Mmakedll
2.2.5. -BstaticCompiles for and links to the static version of the PGI runtime libraries.
Default
The compiler uses static libraries.
Usage
The following command line explicitly compiles for and links to the static version of thePGI runtime libraries:% pgfortran -Bstatic -c object1.f
Command-Line Options Reference
PVF Reference Guide Version 2017 | 13
Description
You can use this option to explicitly compile for and link to the static version of the PGIruntime libraries.
On Windows, -Bstatic must be used for both compiling and linking.
For more information on using static libraries on Windows, refer to ‘Creating and UsingStatic Libraries on Windows’ in the ‘Creating and Using Libraries’ section of the PGICompiler User’s Guide.
Related options
-Bdynamic, -Bstatic_pgi
2.2.6. -Bstatic_pgiLinux only. Compiles for and links to the static version of the PGI runtime libraries.Implies -Mnorpath.
Default
The compiler uses static libraries.
Usage
The following command line explicitly compiles for and links to the static version of thePGI runtime libraries:% pgfortran -Bstatic -c object1.f
Description
You can use this option to explicitly compile for and link to the static version of the PGIruntime libraries.
On Linux, -Bstatic_pgi results in code that runs on most Linux systems withoutrequiring a Portability package.
For more information on using static libraries on Windows, refer to ‘Creating and UsingStatic Libraries on Windows’ in the ‘Creating and Using Libraries’ section of the PVFUser's Guide, https://www.pgroup.com/resources/docs.php.
Related options
-Bdynamic, -Bstatic
Command-Line Options Reference
PVF Reference Guide Version 2017 | 14
2.2.7. -byteswapioSwaps the byte-order of data in unformatted Fortran data files on input/output.
Default
The compiler does not byte-swap data on input/output.
Usage
The following command-line requests that byte-swapping be performed on input/output.$ pgfortran -byteswapio myprog.f
Description
Use the -byteswapio option to swap the byte-order of data in unformatted Fortrandata files on input/output. When this option is used, the order of bytes is swapped inboth the data and record control words; the latter occurs in unformatted sequential files.
You can use this option to convert big-endian format data files produced by most legacyRISC workstations to the little-endian format used on x86-64/x64 or OpenPOWERsystems on the fly during file reads/writes.
This option assumes that the record layouts of unformatted sequential access and directaccess files are the same on the systems. It further assumes that the IEEE representationis used for floating-point numbers. In particular, the format of unformatted data filesproduced by PGI Fortran compilers is identical to the format used on Sun and SGIworkstations; this format allows you to read and write unformatted Fortran datafiles produced on those platforms from a program compiled for an x86-64/x64 orOpenPOWER platform using the -byteswapio option.
Related options
None.
2.2.8. -C(Fortran only) Generates code to check array bounds.
Default
The compiler does not enable array bounds checking.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 15
Usage
In this example, the compiler instruments the executable produced from myprog.f toperform array bounds checking at runtime:$ pgfortran -C myprog.f
Description
Use this option to enable array bounds checking. If an array is an assumed size array,the bounds checking only applies to the lower bound. If an array bounds violationoccurs during execution, an error message describing the error is printed and theprogram terminates. The text of the error message includes the name of the array, thelocation where the error occurred (the source file and the line number in the source), andinformation about the out of bounds subscript (its value, its lower and upper bounds,and its dimension).
Related options
-Mbounds, -Mnobounds
2.2.9. -cHalts the compilation process after the assembling phase and writes the object code to afile.
Default
The compiler produces an executable file and does not use the -c option.
Usage
In this example, the compiler produces the object file myprog.obj in the currentdirectory.$ pgfortran -c myprog.f
Description
Use the -c option to halt the compilation process after the assembling phase and writethe object code to a file. If the input file is filename.f, the output file is .
Related options
-E, -Mkeepasm, -o, -S
Command-Line Options Reference
PVF Reference Guide Version 2017 | 16
2.2.10. -DCreates a preprocessor macro with a given value.
You can use the -D option more than once on a compiler command line. The numberof active macro definitions is limited only by available memory.
Syntax-Dname[=value]
Where name is the symbolic name and value is either an integer value or a characterstring.
Default
If you define a macro name without specifying a value, the preprocessor assigns thestring 1 to the macro name.
Usage
In the following example, the macro PATHLENGTH has the value 256 until asubsequent compilation. If the -D option is not used, PATHLENGTH is set to 128.$ pgfortran -DPATHLENGTH=256 myprog.F
The source text in myprog.F is this: #ifndef PATHLENGTH#define PATHLENGTH 128 #endif SUBROUTINE SUB CHARACTER*PATHLENGTH path ... END
Description
Use the -D option to create a preprocessor macro with a given value. The value must beeither an integer or a character string.
You can use macros with conditional compilation to select source text duringpreprocessing. A macro defined in the compiler invocation remains in effect foreach module on the command line, unless you remove the macro with an #undefpreprocessor directive or with the -U option. The compiler processes all of the -Uoptions in a command line after processing the -D options.
To set this option in PVF, use the Fortran | Preprocessor | Preprocessor Definitionsproperty, described in ‘Preprocessor Definitions’.
Related options
-U
Command-Line Options Reference
PVF Reference Guide Version 2017 | 17
2.2.11. -dryrunDisplays the invocations of the compiler, assembler, and linker but does not executethem.
Default
The compiler does not display individual phase invocations.
Usage
The following command line requests verbose invocation information.$ pgfortran -dryrun myprog.f
Description
Use the -dryrun option to display the invocations of the compiler, assembler, andlinker but not have them executed. These invocations are command lines created by thecompiler driver from the rc files and the command-line supplied with -dryrun.
Related options
-Minfo[=option [,option,...]], -V[release_number], -###
2.2.12. -drystdincDisplays the standard include directories and then exits the compiler.
Default
The compiler does not display standard include directories.
Usage
The following command line requests a display for the standard include directories.$ pgfortran -drystdinc myprog.f
Description
Use the -drystdinc option to display the standard include directories and then exitthe compiler.
Related options
None.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 18
2.2.13. -EHalts the compilation process after the preprocessing phase and displays thepreprocessed output on the standard output.
Default
The compiler produces an executable file.
Usage
In the following example the compiler displays the preprocessed myprog.f on thestandard output.$ pgfortran -E myprog.f
Description
Use the -E option to halt the compilation process after the preprocessing phase anddisplay the preprocessed output on the standard output.
Related options
-C, -c, -Mkeepasm, -o, -F, -S
2.2.14. -FStops compilation after the preprocessing phase.
Default
The compiler produces an executable file.
Usage
In the following example the compiler produces the preprocessed file myprog.f in thecurrent directory.$ pgfortran -F myprog.F
Description
Use the -F option to halt the compilation process after preprocessing and write thepreprocessed output to a file. If the input file is filename.F, then the output file isfilename.f.
Related options
-c, -E, -Mkeepasm, -o, -S
Command-Line Options Reference
PVF Reference Guide Version 2017 | 19
2.2.15. -fastEnables vectorization with SIMD instructions, cache alignment, and flushz for 64-bittargets.
Default
The compiler enables vectorization with SIMD instructions, cache alignment, and flushz.
Usage
In the following example the compiler produces vector SIMD code when targeting a 64-bit machine.$ pgfortran -fast vadd.f95
Description
When you use this option, a generally optimal set of options is chosen for targets thatsupport SIMD capability. In addition, the appropriate -tp option is automaticallyincluded to enable generation of code optimized for the type of system on whichcompilation is performed. This option enables vectorization with SIMD instructions,cache alignment, and flushz.
Auto-selection of the appropriate -tp option means that programs built using the-fastsse option on a given system are not necessarily backward-compatible witholder systems.
C/C++ compilers enable -Mautoinline with -fast.
To set this option in PVF, use the Fortran | General | Optimization property, describedin ‘Optimization’.
Related options
-O<level>, -Munroll[=option [,option...]], -Mnoframe , -Mscalarsse , -M[no]vect[=option [,option,...]], -Mcache_align , -tp <target>[,target...] , -M[no]autoinline[=option[,option,...]]
2.2.16. -fastsseSynonymous with -fast.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 20
2.2.17. --flagcheckCauses the compiler to check that flags are correct and then exit without anycompilation occuring.
Default
The compiler begins a compile without the additional step to first validate that flags arecorrect.
Usage
In the following example the compiler checks that flags are correct, and then exits.$ pgfortran --flagcheck myprog.f
Description
Use this option to make the compiler check that flags are correct and then exit. If flagsare all correct then the compiler returns a zero status. No compilation occurs.
Related options
None.
2.2.18. -flagsDisplays valid driver options on the standard output.
Default
The compiler does not display the driver options.
Usage
In the following example the user requests information about the known switches.$ pgfortran -flags
Description
Use this option to display driver options on the standard output. When you use thisoption with -v, in addition to the valid options, the compiler lists options that arerecognized and ignored.
Related options
-#, -###, -v
Command-Line Options Reference
PVF Reference Guide Version 2017 | 21
2.2.19. -gInstructs the compiler to include symbolic debugging information in the object module;sets the optimization level to zero unless a -O option is present on the command line.
Default
The compiler does not put debugging information into the object module.
Usage
In the following example, the object file myprog.obj contains symbolic debugginginformation.$ pgfortran -c -g myprog.f
Description
Use the -g option to instruct the compiler to include symbolic debugging information inthe object module. Debuggers, including the PGI debugger, require symbolic debugginginformation in the object module to display and manipulate program variables andsource code.
If you specify the -g option on the command-line, the compiler sets the optimizationlevel to -O0 (zero), unless you specify the -O option. For more information on theinteraction between the -g and -O options, refer to the -O entry. Symbolic debuggingmay give confusing results if an optimization level other than zero is selected.
Including symbolic debugging information increases the size of the object module.
To set this option in PVF, use the Fortran | General | Debug Information Formatproperty, described in ‘Debug Information Format’ on page 377.
Related options
-O<level>, -gopt
2.2.20. -goptInstructs the compiler to include symbolic debugging information in the object file, andto generate optimized code identical to that generated when -g is not specified.
Default
The compiler does not put debugging information into the object module.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 22
Usage
In the following example, the object file myprog.obj contains symbolic debugginginformation.$ pgfortran -c -gopt myprog.f
Description
Using -g alters how optimized code is generated in ways that are intended to enableor improve debugging of optimized code. The -gopt option instructs the compiler toinclude symbolic debugging information in the object file, and to generate optimizedcode identical to that generated when -g is not specified.
To set this option in PVF, use the Fortran | General | Debug Information Formatproperty described in ‘Debug Information Format’.
Related options
-g, -M<pgflag>
2.2.21. -helpUsed with no other options, -help displays options recognized by the driver on thestandard output. When used in combination with one or more additional options, usageinformation for those options is displayed to standard output.
Default
The compiler does not display usage information.
Usage
In the following example, usage information for -Minline is printed to standardoutput.$ pgcc -help -Minline -Minline[=lib:<inlib>|<maxsize>|<func>|except:<func>|name:<func>|maxsize:<n>|totalsize:<n>|smallsize:<n>|reshape] Enable function inlining lib:<inlib> Use extracted functions from inlib <maxsize> Set maximum function size to inline <func> Inline function func except:<func> Do not inline function func name:<func> Inline function func maxsize:<n> Inline only functions smaller than n totalsize:<n> Limit inlining to total size of n smallsize:<n> Always inline functions smaller than n reshape Allow inlining in Fortran even when array shapes do not match -Minline Inline all functions that were extracted
Command-Line Options Reference
PVF Reference Guide Version 2017 | 23
In the following example, usage information for -help shows how groups of optionscan be listed or examined according to function.$ pgcc -help -help -help[=groups|asm|debug|language|linker|opt|other|overall|phase|prepro|suffix|switch|target|variable]
Description
Use the -help option to obtain information about available options and their syntax. Youcan use -help in one of three ways:
‣ Use -help with no parameters to obtain a list of all the available options with abrief one-line description of each.
‣ Add a parameter to -help to restrict the output to information about a specificoption. The syntax for this usage is this:-help <command line option>
‣ Add a parameter to -help to restrict the output to a specific set of options or to abuilding process. The syntax for this usage is this:-help=<subgroup>
The following table lists and describes the subgroups available with -help.
Table 9 Subgroups for -help Option
Use this -help option To get this information...
-help=asm A list of options specific to the assembly phase.
-help=debug A list of options related to debug information generation.
-help=groups A list of available switch classifications.
-help=language A list of language-specific options.
-help=linker A list of options specific to link phase.
-help=opt A list of options specific to optimization phase.
-help=other A list of other options, such as ANSI conformance pointer aliasing for C.
-help=overall A list of options generic to any PGI compiler.
-help=phase A list of build process phases and to which compiler they apply.
-help=prepro A list of options specific to the preprocessing phase.
-help=suffix A list of known file suffixes and to which phases they apply.
-help=switch A list of all known options; this is equivalent to usage of -help without any
parameter.
-help=target A list of options specific to target processor.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 24
Use this -help option To get this information...
-help=variable A list of all variables and their current value. They can be redefined on the
command line using syntax VAR=VALUE.
Related options
-#, -###, -show, -V[release_number], -flags
2.2.22. -IAdds a directory to the search path for files that are included using either the INCLUDEstatement or the preprocessor directive #include.
Default
The compiler searches only certain directories for included files.
Syntax-Idirectory
Where directory is the name of the directory added to the standard search path forinclude files.
Usage
In the following example, the compiler first searches the directory mydir and thensearches the default directories for include files.$ pgfortran -Imydir
Description
Adds a directory to the search path for files that are included using the INCLUDEstatement or the preprocessor directive #include. Use the -I option to add a directoryto the list of where to search for the included files. The compiler searches the directoryspecified by the -I option before the default directories.
The Fortran INCLUDE statement directs the compiler to begin reading from another file.The compiler uses two rules to locate the file:
‣ If the file name specified in the INCLUDE statement includes a path name, thecompiler begins reading from the file it specifies.
‣ If no path name is provided in the INCLUDE statement, the compiler searches (inorder):
1. Any directories specified using the -I option (in the order specified) 2. The directory containing the source file 3. The current directory
Command-Line Options Reference
PVF Reference Guide Version 2017 | 25
For example, the compiler applies rule (1) to the following statements:INCLUDE '/bob/include/file1' (absolute path name)INCLUDE '../../file1' (relative path name)
and rule (2) to this statement:INCLUDE 'file1'
To set this option in PVF, use the Fortran | General | Additional Include Directoriesproperty, described in ‘Additional Include Directories’, or the Fortran | Preprocessor |Additional Include Directories property, described in ‘Additional Include Directories’.
Related options
-Mnostdinc
2.2.23. -i2, -i4, -i8Treat INTEGER and LOGICAL variables as either two, four, or eight bytes.
Default
The compiler treats INTERGER and LOGICAL variables as four bytes.
Usage
In the following example, using the -i8 switch causes the integer variables to be treatedas 64 bits.$ pgfortran -i8 int.f
int.f is a function similar to this:int.f print *, "Integer size:", bit_size(i) end
Description
Use this option to treat INTEGER and LOGICAL variables as either two, four, or eightbytes. INTEGER*8 values not only occupy 8 bytes of storage, but operations use 64 bits,instead of 32 bits.
‣ -i2: Treat INTEGER variables as 2 bytes.‣ -i4: Treat INTEGER variables as 4 bytes.‣ -i8: Treat INTEGER and LOGICAL variables as 8 bytes and use 64-bits for
INTEGER*8 operations.
Related options
None.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 26
2.2.24. -K<flag>Requests that the compiler provide special compilation semantics with regard toconformance to IEEE 754.
Default
The default is -Knoieee and the compiler does not provide special compilationsemantics.
Syntax
-K<flag>
Where flag is one of the following:
ieee Perform floating-point operations in strict conformance with the IEEE 754 standard.
Some optimizations are disabled, and on some systems a more accurate math library is
linked if -Kieee is used during the link step.
To set this option in PVF, use the Fortran | Floating Point Options | IEEE Arithmetic
property, described in ‘IEEE Arithmetic’.
noieee Default flag. Use the fastest available means to perform floating-point operations, link
in faster non-IEEE libraries if available, and disable underflow traps.
trap=option
[,option]...
Controls the behavior of the processor when floating-point exceptions occur.
Possible options include:
fp
align (ignored)
inv
denorm
divz
ovf
unf
inexact
Usage
In the following example, the compiler performs floating-point operations in strictconformance with the IEEE 754 standard$ pgfortran -Kieee myprog.f
Description
Use -K to instruct the compiler to provide special compilation semantics.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 27
-Ktrap is only processed by the compilers when compiling main functions orprograms. The options inv, denorm, divz, ovf, unf, and inexact correspond to theprocessor’s exception mask bits: invalid operation, denormalized operand, divide-by-zero, overflow, underflow, and precision, respectively.
Normally, the processor’s exception mask bits are on, meaning that floating-pointexceptions are masked – the processor recovers from the exceptions and continues. Ifa floating-point exception occurs and its corresponding mask bit is off, or "unmasked",execution terminates with an arithmetic exception (C's SIGFPE signal).
-Ktrap=fp is equivalent to -Ktrap=inv,divz,ovf.
To set this option in PVF, use the Fortran | Floating Point Options | Floating PointException Handling property, described in ‘Floating Point Exception Handling’.
The PGI compilers do not support exception-free execution for -Ktrap=inexact.The purpose of this hardware support is for those who have specific uses for itsexecution, along with the appropriate signal handlers for handling exceptions itproduces. It is not designed for normal floating point operation code support.
Related options
None.
2.2.25. --keeplnk(Windows only.) Preserves the temporary file when the compiler generates a temporaryindirect file for a long linker command.
Usage
In the following example the compiler preserves each temporary file rather than deletingit.$ pgfortran --keeplnk myprog.f
Description
If the compiler generates a temporary indirect file for a long linker command, use thisoption to instruct the compiler to preserve the temporary file instead of deleting it.
Related options
None.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 28
2.2.26. -LSpecifies a directory to search for libraries.
Multiple -L options are valid. However, the position of multiple -L options isimportant relative to -l options supplied.
Default
The compiler searches the standard library directory.
Syntax-Ldirectory
Where directory is the name of the library directory.
Usage
In the following example, the library directory is /lib and the linker links in thestandard libraries required by PGFORTRAN from this directory.$ pgfortran -L/lib myprog.f
In the following example, the library directory /lib is searched for the library filelibx.a and both the directories /lib and /libz are searched for liby.a.$ pgfortran -L/lib -lx -L/libz -ly myprog.f
Description
Use the -L option to specify a directory to search for libraries. Using -L allows you to adddirectories to the search path for library files.
Related options
-I
2.2.27. -l<library>Instructs the linker to load the specified library. The linker searches <library>in additionto the standard libraries.
The linker searches the libraries specified with -l in order of appearance beforesearching the standard libraries.
Syntax-llibrary
Command-Line Options Reference
PVF Reference Guide Version 2017 | 29
Where library is the name of the library to search.
Usage: In the following example, if the standard library directory is /lib the linkerloads the library /lib/libmylib.a, in addition to the standard libraries.$ pgfortran myprog.f -lmylib
Description
Use this option to instruct the linker to load the specified library. The compiler prependsthe characters lib to the library name and adds the .a extension following the libraryname. The linker searches each library specified before searching the standard libraries.
Related options
-L
2.2.28. -MGenerate make dependence lists. You can use -MD,filename (pgc++ only) to generatemake dependence lists and print them to the specified file.
2.2.29. -mDisplays a link map on the standard output.
Default
The compiler does display the link map and does not use the -m option.
Usage
When the following example is executed on Windows, pgfortran creates a link map inthe file myprog.map.$ pgfortran -m myprog.f
Description
Use this option to display a link map.
‣ On Linux, the map is written to stdout.‣ On Windows, the map is written to a .map file whose name depends on the
executable. If the executable is myprog.f, the map file is in myprog.map.
Related options
-c, -o, -u
Command-Line Options Reference
PVF Reference Guide Version 2017 | 30
2.2.30. -m64Use the 64-bit compiler for the default processor type.
Usage
When the following example is executed, pgfortran uses the 64-bit compiler for thedefault processor type.$ pgfortran -m64 myprog.f
Description
Use this option to specify the 64-bit compiler as the default processor type.
2.2.31. -M<pgflag>Selects options for code generation. The options are divided into the followingcategories:
Code generation Fortran Language Controls Optimization
Environment C/C++ Language Controls Miscellaneous
Inlining
The following table lists and briefly describes the options alphabetically and includesa field showing the category. For more details about the options as they relate to thesecategories, refer to ‘-M Options by Category’ on page 113.
Table 10 -M Options Summary
pgflag Description Category
allocatable=95|03 Controls whether to use Fortran 95 or Fortran 2003
semantics in allocatable array assignments.
Fortran Language
anno Annotate the assembly code with source code. Miscellaneous
[no]autoinline When a C/C++ function is declared with the inline
keyword, inline it at -O2.
Inlining
[no]backslash Determines how the backslash character is treated in
quoted strings.
Fortran Language
[no]bounds Specifies whether array bounds checking is enabled or
disabled.
Miscellaneous
byteswapio Swap byte-order (big-endian to little-endian or vice
versa) during I/O of Fortran unformatted data.
Miscellaneous
Command-Line Options Reference
PVF Reference Guide Version 2017 | 31
pgflag Description Category
cache_align Where possible, align data objects of size greater than
or equal to 16 bytes on cache-line boundaries.
Optimization
chkfpstk Check for internal consistency of the x87 FP stack in
the prologue of a function and after returning from a
function or subroutine call (-tp px/p5/p6/piii targets
only).
Miscellaneous
chkptr Check for NULL pointers (pgf95, pgfortran only). Miscellaneous
chkstk Check the stack for available space upon entry to and
before the start of a parallel region. Useful when many
private variables are declared.
Miscellaneous
concur Enable auto-concurrentization of loops. Multiple
processors or cores will be used to execute
parallelizable loops.
Optimization
cpp Run the PGI cpp-like preprocessor without performing
subsequent compilation steps.
Miscellaneous
cray Force Cray Fortran (CF77) compatibility. Optimization
cuda Enables CUDA Fortran. Fortran Language
[no]daz Do/don’t treat denormalized numbers as zero. Code Generation
[no]dclchk Determines whether all program variables must be
declared.
Fortran Language
[no]defaultunit Determines how the asterisk character ("*") is treated
in relation to standard input and standard output,
regardless of the status of I/O units 5 and 6..
Fortran Language
[no]depchk Checks for potential data dependencies. Optimization
[no]dse Enables [disables] dead store elimination phase for
programs making extensive use of function inlining.
Optimization
[no]dlines Determines whether the compiler treats lines
containing the letter "D" in column one as executable
statements.
Fortran Language
dll Link with the DLL version of the runtime libraries
(Windows only).
Miscellaneous
dollar,char Specifies the character to which the compiler maps the
dollar sign code.
Fortran Language
[no]dwarf Specifies not to add DWARF debug information. Code Generation
Command-Line Options Reference
PVF Reference Guide Version 2017 | 32
pgflag Description Category
dwarf1 When used with -g, generate DWARF1 format debug
information.
Code Generation
dwarf2 When used with -g, generate DWARF2 format debug
information.
Code Generation
dwarf3 When used with -g, generate DWARF3 format debug
information.
Code Generation
extend Instructs the compiler to accept 132-column source
code; otherwise it accepts 72-column code.
Fortran Language
extract invokes the function extractor. Inlining
[no]f[=option] Perform certain floating point intrinsic functions using
relaxed precision.
Optimization
fixed Instructs the compiler to assume F77-style fixed format
source code (pgf95, pgfortran only).
Fortran Language
[no]flushz Do/don't set SSE flush-to-zero mode Code Generation
[no]fpapprox Specifies not to use low-precision fp approximation
operations.
Optimization
free Instructs the compiler to assume F90-style free format
source code.
Fortran Language
func32 The compiler aligns all functions to 32-byte
boundaries.
Code Generation
gccbug[s] Matches behavior of certain gcc bugs Miscellaneous
info Prints informational messages regarding optimization
and code generation to standard output as compilation
proceeds.
Miscellaneous
inform Specifies the minimum level of error severity that the
compiler displays.
Miscellaneous
inline Invokes the function inliner. Inlining
[no]iomutex Determines whether critical sections are generated
around Fortran I/O calls.
Fortran Language
[no]ipa Invokes interprocedural analysis and optimization. Optimization
keepasm Instructs the compiler to keep the assembly file. Miscellaneous
largeaddressaware [Win64 only] Generates code that allows for addresses
greater than 2GB, using RIP-relative addressing.
Code Generation
Command-Line Options Reference
PVF Reference Guide Version 2017 | 33
pgflag Description Category
[no]large_arrays Enables support for 64-bit indexing and single static
data objects of size larger than 2GB.
Code Generation
list Specifies whether the compiler creates a listing file. Miscellaneous
[no]loop32 Aligns [does not align] innermost loops on 32 byte
boundaries with -tp barcelona
Code Generation
[no]lre Enable [disable] loop-carried redundancy elimination. Optimization
makedll Generate a dynamic link library (DLL).. Miscellaneous
makeimplib Passes the -def switch to the librarian without a
deffile, when used without -def:deffile.
Miscellaneous
mpi=option Link to MPI libraries: MPICH, SGI, or Microsoft MPI
libraries
Code Generation
neginfo Instructs the compiler to produce information on why
certain optimizations are not performed.
Miscellaneous
noframe Eliminates operations that set up a true stack frame
pointer for functions.
Optimization
noi4 Determines how the compiler treats INTEGER
variables.
Optimization
nomain When the link step is called, don’t include the object
file that calls the Fortran main program..
Code Generation
noopenmp When used in combination with the -mp option, the
compiler ignores OpenMP parallelization directives ,
but still processes SGI-style parallelization directives.
Miscellaneous
nopgdllmain Do not link the module containing the default DllMain()
into the DLL.
Miscellaneous
nosgimp When used in combination with the -mp option, the
compiler ignores SGI-style parallelization directives,
but still processes OpenMP directives.
Miscellaneous
nostdinc Instructs the compiler to not search the standard
location for include files. To set this option in PVF, use
the Fortran | Preprocessor | Ignore Standard Include
Path property.
Environment
nostdlib Instructs the linker to not link in the standard libraries. Environment
[no]onetrip Determines whether each DO loop executes at least
once.
Language
Command-Line Options Reference
PVF Reference Guide Version 2017 | 34
pgflag Description Category
novintr Disable idiom recognition and generation of calls to
optimized vector functions.
Optimization
pfi Instrument the generated code and link in libraries for
dynamic collection of profile and data information at
runtime.
Optimization
pre Read a pgfi.out trace file and use the information to
enable or guide optimizations.
Optimization
[no]pre Force [disable] generation of non-temporal moves and
prefetching.
Code Generation
[no]prefetch Enable [disable] generation of prefetch instructions. Optimization
preprocess Perform cpp-like preprocessing on assembly language
and Fortran input source files.
Miscellaneous
prof Enable Compiler feedback and modify DWARF sections. Code Generation
[no]r8 Determines whether the compiler promotes REAL
variables and constants to DOUBLE PRECISION.
Optimization
[no]r8intrinsics Determines how the compiler treats the intrinsics
CMPLX and REAL.
Optimization
[no]recursive Allocate [do not allocate] local variables on the stack;
this allows recursion. SAVEd, data-initialized, or
namelist members are always allocated statically,
regardless of the setting of this switch.
Code Generation
[no]reentrant Specifies whether the compiler avoids optimizations
that can prevent code from being reentrant.
Code Generation
[no]ref_externals Do [do not] force references to names appearing in
EXTERNAL statements.
Code Generation
safe_lastval In the case where a scalar is used after a loop, but
is not defined on every iteration of the loop, the
compiler does not by default parallelize the loop.
However, this option tells the compiler it is safe to
parallelize the loop. For a given loop, the last value
computed for all scalars make it safe to parallelize the
loop.
Code Generation
[no]save Determines whether the compiler assumes that all
local variables are subject to the SAVE statement.
Fortran Language
Command-Line Options Reference
PVF Reference Guide Version 2017 | 35
pgflag Description Category
[no]scalarsse Do [do not] use SSE/SSE2 instructions to perform scalar
floating-point arithmetic.
Optimization
[no]second_underscore Do [do not] add the second underscore to the name
of a Fortran global if its name already contains an
underscore.
Code Generation
[no]signextend Do [do not] extend the sign bit, if it is set. Code Generation
[no]smart Do [do not] enable optional post-pass assembly
optimizer.
Optimization
[no]smartalloc[=huge|
huge:<n>|hugebss]
Add a call to the routine mallopt in the main routine.
Supports large TLBs on Linux and Windows.
Tip To be effective, this switch mustbe specified when compiling the filecontaining the Fortran, C, or C++ mainprogram.
Environment
standard Causes the compiler to flag source code that does not
conform to the ANSI standard.
Fortran Language
[no]stride0 Do [do not] generate alternate code for a loop that
contains an induction variable whose increment may
be zero.
Code Generation
[no]unixlogical Determines how the compiler treats logical values.. Fortran Language
[no]unroll Controls loop unrolling. Optimization
[no]upcase Determines whether the compiler preserves uppercase
letters in identifiers..
Fortran Language
varargs Forces Fortran program units to assume calls are to C
functions with a varargs type interface .
Code Generation
[no]vect Do [do not] invoke the code vectorizer. Optimization
2.2.32. -module <moduledir>Allows you to specify a particular directory in which generated intermediate .mod filesshould be placed.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 36
Default
The compiler places .mod files in the current working directory, and searches only in thecurrent working directory for pre-compiled intermediate .mod files.
Usage
The following command line requests that any intermediate module file producedduring compilation of myprog.f be placed in the directory mymods; specifically, the file./mymods/myprog.mod is used.$ pgfortran -module mymods myprog.f
Description
Use the -module option to specify a particular directory in which generatedintermediate .mod files should be placed. If the -module <moduledir> option ispresent, and USE statements are present in a compiled program unit, then <moduledir> issearched for .mod intermediate files prior to a search in the default local directory.
To set this option in PVF, use the Fortran | Output | Module Path property, described in‘Module Path’.
Related options
None.
2.2.33. -mpInstructs the compiler to interpret user-inserted OpenMP shared-memory parallelprogramming directives, and to generate an executable file which will utilize multipleprocessors in a shared-memory parallel system.
Default
The compiler interprets user-inserted shared-memory parallel programming directiveswhen linking. To disable this option, use the -nomp option when linking.
Usage
The following command line requests processing of any shared-memory directivespresent in myprog.f:$ pgfortran -mp myprog.f
Description
Use the -mp option to instruct the compiler to interpret user-inserted OpenMP shared-memory parallel programming directives and to generate an executable file whichutilizes multiple processors in a shared-memory parallel system.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 37
The suboptions are one or more of the following:align
Forces loop iterations to be allocated to OpenMP processes using an algorithmthat maximizes alignment of vector sub-sections in loops that are both parallelizedand vectorized for SSE. This allocation can improve performance in program unitsthat include many such loops. It can also result in load-balancing problems thatsignificantly decrease performance in program units with relatively short loops thatcontain a large amount of work in each iteration. The numa suboption uses libnumaon systems where it is available.
allcoresInstructs the compiler to target all available cores. You specify this suboption at linktime.
bindInstructs the compiler to bind threads to cores. You specify this suboption at linktime.
[no]numaUses [does not use] libnuma on systems where it is available.
For a detailed description of this programming model and the associated directives,refer to Section 9, ‘Using OpenMP’ of the PGI Compiler User's Guide.
To set this option in PVF, use the Fortran | Language | Enable OpenMP Directivesproperty, described in ‘Enable OpenMP Directives’.
Related options
-Mconcur[=option [,option,...]], -M[no]vect[=option [,option,...]]
2.2.34. -noswitcherrorIssues warnings instead of errors for unknown switches. Ignores unknown commandline switches after printing a warning message.
Default
The compiler prints an error message and then halts.
Usage
In the following example, the compiler ignores unknown command line switches afterprinting a warning message.$ pgfortran -noswitcherror myprog.f
Command-Line Options Reference
PVF Reference Guide Version 2017 | 38
Description
Use this option to instruct the compiler to ignore unknown command line switches afterprinting an warning message.
Tip You can configure this behavior in the siterc file by adding: setNOSWITCHERROR=1.
Related options
None.
2.2.35. -O<level>Invokes code optimization at the specified level.
Default
The compiler optimizes at level 2.
Syntax-O [level]
Where level is an integer from 0 to 4.
Usage
In the following example, since no -O option is specified, the compiler sets theoptimization to level 1.$ pgfortran myprog.f
In the following example, since no optimization level is specified and a -O option isspecified, the compiler sets the optimization to level 2.$ pgfortran -O myprog.f
Description
Use this option to invoke code optimization.Using the PGI compiler commands withthe -Olevel option (the capital O is for Optimize), you can specify any of the followingoptimization levels:-O0
Level zero specifies no optimization. A basic block is generated for each languagestatement.
-O1Level one specifies local optimization. Scheduling of basic blocks is performed.Register allocation is performed.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 39
-OWhen no level is specified, level two global optimizations are performed, includingtraditional scalar optimizations, induction recognition, and loop invariant motion. NoSIMD vectorization is enabled.
-O2Level two specifies global optimization. This level performs all level-one localoptimization as well as level-two global optimization described in -O. In addition,this level enables more advanced optimizations such as SIMD code generation, cachealignment, and partial redundancy elimination.
-O3Level three specifies aggressive global optimization. This level performs all level-one and level-two optimizations and enables more aggressive hoisting and scalarreplacement optimizations that may or may not be profitable.
-O4Level four performs all level-one, level-two, and level-three optimizations andenables hoisting of guarded invariant floating point expressions.
To set this option (-O2 or -O3) in PVF, use the Fortran | Optimization | GlobalOptimizations property, described in ‘Global Optimizations’.
The following table shows the interaction between the -O option, -g option, -Mvect,and -Mconcur options.
Table 11 Optimization and -O, -g, -Mvect, and -Mconcur Options
Optimize Option Debug Option -M Option Optimization Level
none none none 1
none none -Mvect 2
none none -Mconcur 2
none -g none 0
-O none or -g none 2
-Olevel none or -g none level
-Olevel < 2 none or -g -Mvect 2
-Olevel < 2 none or -g -Mconcur 2
Unoptimized code compiled using the option -O0 can be significantly slower than codegenerated at other optimization levels. Like the -Mvect option, the -Munroll optionsets the optimization level to level-2 if no -O or -g options are supplied. The -goptoption is recommended for generation of debug information with optimized code. Formore information on optimization, refer to the ‘Optimizing and Parallelizing’ section ofthe PVF User's Guide, https://www.pgroup.com/resources/docs.php.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 40
Related options
-g, -M<pgflag>, -gopt
2.2.36. -oNames the executable file. Use the -o option to specify the filename of the compilerobject file. The final output is the result of linking.
Default
The compiler creates executable filenames as needed. If you do not specify the -ooption, the default filename is the linker output file with a name comprised of the basefile name, such as myprog, plus the extension .exe, for example: myprog.exe .
Syntax
-o filename
Where filename is the name of the file for the compilation output. The filename should nothave a .f extension.
Usage
In the following example, the executable file is myp.exe instead of the default a.outmyprog.exe.$ pgfortran myprog.f -o myp
To set this option in PVF, use the Fortran | Output | Object File Name property,described in ‘Object File Name’ on page 377.
Related options
-c, -E, -F, -S
2.2.37. -pc
This option is available only for -tp px/p5/p6/piii targets.
Allows you to control the precision of operations performed using the x87 floating pointunit, and their representation on the x87 floating point stack.
Syntax-pc { 32 | 64 | 80 }
Usage$ pgfortran -pc 64 myprog.f
Command-Line Options Reference
PVF Reference Guide Version 2017 | 41
Description
The x87 architecture implements a floating-point stack using eight 80-bit registers. Eachregister uses bits 0–63 as the significant, bits 64–78 for the exponent, and bit 79 is the signbit. This 80-bit real format is the default format, called the extended format. When valuesare loaded into the floating point stack they are automatically converted into extendedreal format. The precision of the floating point stack can be controlled, however, bysetting the precision control bits (bits 8 and 9) of the floating control word appropriately.In this way, you can explicitly set the precision to standard IEEE double-precision using64 bits, or to single precision using 32 bits.
According to Intel documentation, this only affects the x87 operations of add, subtract,multiply, divide, and square root. In particular, it does not appear to affect the x87transcendental instructions.
The default precision is system dependent. To alter the precision in a given programunit, the main program must be compiled with the same -pc option. The command lineoption -pc val lets the programmer set the compiler’s precision preference.
Valid values for val are:
32 single precision 64 double precision 80 extended precision
Extended Precision Option – Operations performed exclusively on the floating-pointstack using extended precision, without storing into or loading from memory, can causeproblems with accumulated values within the extra 16 bits of extended precision values.This can lead to answers, when rounded, that do not match expected results.
For example, if the argument to sin is the result of previous calculations performed onthe floating-point stack, then an 80-bit value used instead of a 64-bit value can result inslight discrepancies. Results can even change sign due to the sin curve being too closeto an x-intercept value when evaluated. To maintain consistency in this case, you canassure that the compiler generates code that calls a function. According to the x86 ABI, afunction call must push its arguments on the stack (in this way memory is guaranteed tobe accessed, even if the argument is an actual constant). Thus, even if the called functionsimply performs the inline expansion, using the function call as a wrapper to sin hasthe effect of trimming the argument precision down to the expected size. Using the-Mnobuiltin option on the command line for C accomplishes this task by resolvingall math routines in the library libm, performing a function call of necessity. The othermethod of generating a function call for math routines, but one that may still producethe inline instructions, is by using the -Kieee switch.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 42
A second example illustrates the precision control problem using a section of code todetermine machine precision:program find_precision w = 1.0 100 w=w+w y=w+1 z=y-w if (z .gt. 0) goto 100 C now w is just big enough that |((w+1)-w)-1| >= 1 ... print*,w end
In this case, where the variables are implicitly real*4, operations are performed on thefloating-point stack where optimization removes unnecessary loads and stores frommemory. The general case of copy propagation being performed follows this pattern:a = x y = 2.0 + a
Instead of storing x into a, then loading a to perform the addition, the value of x canbe left on the floating-point stack and added to 2.0. Thus, memory accesses in somecases can be avoided, leaving answers in the extended real format. If copy propagationis disabled, stores of all left-hand sides will be performed automatically and reloadedwhen needed. This will have the effect of rounding any results to their declared sizes.
The find_precision program has a value of 1.8446744E+19 when executed usingdefault (extended) precision. If, however, -Kieee is set, the value becomes 1.6777216E+07 (single precision.) This difference is due to the fact that -Kieee disables copypropagation, so all intermediate results are stored into memory, then reloaded whenneeded. Copy propagation is only disabled for floating-point operations, not integer.With this particular example, setting the -pc switch will also adjust the result.
The -Kieee switch also has the effect of making function calls to perform alltranscendental operations. Except when the -Mnobuiltin switch is set in C, thefunction still produces the x86 machine instruction for computation, and arguments arepassed on the stack, which results in a memory store and load.
Finally, -Kieee also disables reciprocal division for constant divisors. That is, for a/bwith unknown a and constant b, the expression is usually converted at compile timeto a*(1/b), thus turning an expensive divide into a relatively fast scalar multiplication.However, numerical discrepancies can occur when this optimization is used.
Understanding and correctly using the -pc, -Mnobuiltin, and -Kieee switchesshould enable you to produce the desired and expected precision for calculations whichutilize floating-point operations.
Related options
-K<flag>, Mnobuiltin
Command-Line Options Reference
PVF Reference Guide Version 2017 | 43
2.2.38. --pedanticPrints warnings from included <system header files>.
Default
The compiler prints the warnings from the included system header files.
Usage
In the following example, the compiler prints the warnings from the included systemheader files.$ pgc++ --power myprog.cc
Related options
None.
2.2.39. -pgc++libsInstructs the compiler to append C++ runtime libraries to the link line for programsbuilt using either PGF77 or PGF90 .
Default
The C/C++ compilers do not append the C++ runtime libraries to the link line.
Usage
In the following example the C++ runtime libraries are linked with an object filecompiled with pgf77 .
$ pgf90 main.f90 mycpp.o -pgc++libs
Description
Use this option to instruct the compiler to append C++ runtime libraries to the link linefor programs built using either PGF77 or PGF90 .
Related options
-pgf90libs , -pgf77libs
2.2.40. -pgf77libsInstructs the compiler to append PGF77 runtime libraries to the link line.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 44
Default
The C/C++ compilers do not append the PGF77 runtime libraries to the link line.
Usage
In the following example a .c main program is linked with an object file compiled withpgf77.$ pgcc main.c myf77.o -pgf77libs
Description
Use this option to instruct the compiler to append PGF77 runtime libraries to the linkline.
Related options
-pgc++libs, -pgf90libs
2.2.41. -pgf90libsInstructs the compiler to append PGF90/PGF95/PGFORTRAN runtime libraries to thelink line.
Default
The C/C++ compilers do not append the PGF90/PGF95/PGFORTRAN runtime librariesto the link line.
Usage
In the following example a .c main program is linked with an object file compiled withpgfortran.$ pgcc main.c myf95.o -pgf90libs
Description
Use this option to instruct the compiler to append PGF90/PGF95/PGFORTRAN runtimelibraries to the link line.
Related options
-pgc++libs , -pgf77libs
2.2.42. -r4 and -r8Interprets DOUBLE PRECISION variables as REAL (-r4), or interprets REAL variables asDOUBLE PRECISION (-r8).
Command-Line Options Reference
PVF Reference Guide Version 2017 | 45
Usage
In this example, the double precision variables are interpreted as REAL.$ pgfortran -r4 myprog.f
Description
Interpret DOUBLE PRECISION variables as REAL (-r4) or REAL variables as DOUBLEPRECISION (-r8).
Related options
-i2, -i4, -i8, -Mnor8
2.2.43. -rcSpecifies the name of the driver startup configuration file. If the file or pathnamesupplied is not a full pathname, the path for the configuration file loaded is relativeto the $DRIVER path (the path of the currently executing driver). If a full pathname issupplied, that file is used for the driver configuration file.
Syntax-rc [path] filename
Where path is either a relative pathname, relative to the value of $DRIVER, or a fullpathname beginning with "/". Filename is the driver configuration file.
Usage
In the following example, the file .pgfortranrctest, relative to /usr/pgi/linux86-64/bin , the value of $DRIVER, is the driver configuration file.$ pgfortran -rc .pgfortranrctest myprog.f
Description
Use this option to specify the name of the driver startup configuration file. If the fileor pathname supplied is not a full pathname, the path for the configuration file loadedis relative to the $DRIVER path – the path of the currently executing driver. If a fullpathname is supplied, that file is used for the driver configuration file.
Related options
-show
2.2.44. -SStops compilation after the compiling phase and writes the assembly-language output toa file.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 46
Default
The compiler does not retain a .s file.
Usage
In this example, pgfortran produces the file myprog.s in the current directory.$ pgfortran -S myprog.f
Description
Use this option to stop compilation after the compiling phase and then write theassembly-language output to a file. If the input file is filename.f, then the output fileis filename.s.
Related options
-c, -E, -F, -Mkeepasm, -o
2.2.45. -showProduces driver help information describing the current driver configuration.
Default
The compiler does not show driver help information.
Usage
In the following example, the driver displays configuration information to the standardoutput after processing the driver configuration file.$ pgfortran -show myprog.f
Description
Use this option to produce driver help information describing the current driverconfiguration.
Related options
-V[release_number], -v, -###, -help, -rc
2.2.46. -silentDo not print warning messages.
Default
The compiler prints warning messages.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 47
Usage
In the following example, the driver does not display warning messages.$ pgfortran -silent myprog.f
Description
Use this option to suppress warning messages.
Related options
-v, -V[release_number], -w
2.2.47. -stack(Windows only) Allows you to explicitly set stack properties for your program.
Default
If -stack is not specified, then the defaults are as followed:Win64
No default setting
Syntax-stack={ (reserved bytes)[,(committed bytes)] }{, [no]check }
Usage
The following example demonstrates how to reserve 524,288 stack bytes (512KB),commit 262,144 stack bytes for each routine (256KB), and disable the stack initializationcode with the nocheck argument.$ pgfortran -stack=524288,262144,nocheck myprog.f
Description
Use this option to explicitly set stack properties for your program. The -stack optiontakes one or more arguments: (reserved bytes), (committed bytes), [no]check.reserved bytes
Specifies the total stack bytes required in your program.committed bytes
Specifies the number of stack bytes that the Operating System will allocate for eachroutine in your program. This value must be less than or equal to the stack reservedbytes value.
Default for this argument is 4096 bytes.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 48
[no]checkInstructs the compiler to generate or not to generate stack initialization code uponentry of each routine. Check is the default, so stack initialization code is generated.
Stack initialization code is required when a routine's stack exceeds the committed bytessize. When your committed bytes is equal to the reserved bytes or equal to the stack bytesrequired for each routine, then you can turn off the stack initialization code using the-stack=nocheck compiler option. If you do this, the compiler assumes that you arespecifying enough committed stack space; and therefore, your program does not have tomanage its own stack size.
For more information on determining the amount of stack required by your program,refer to -Mchkstk compiler option, described in ‘Miscellaneous Controls’.
-stack=(reserved bytes),(committed bytes) are linker options.
-stack=[no]check is a compiler option.
If you specify -stack=(reserved bytes),(committed bytes) onyour compile line, it is only used during the link step of your build. Similarly, -stack=[no]check can be specified on your link line, but it's only used during thecompile step of your build.
Related options
-Mchkstk
2.2.48. -ta=tesla(tesla_suboptions),hostDefines the target accelerator and the type of code to generate. This flag is valid forFortran, C, and C++ on supported platforms.
There are three major suboptions:
tesla(:tesla_suboptions)
host
multicore
Default
The compiler uses -ta=tesla,host.
Usage
In the following example, tesla is the accelerator target architecture and the acceleratorgenerates code for compute capability 3.0.$ pgfortran -ta=tesla,cc30
Command-Line Options Reference
PVF Reference Guide Version 2017 | 49
Description
Use this option to select the accelerator target and, optionally, to define the type of codeto genertate.
The -ta flag has the following options:tesla
Select the tesla accelerator target. This option has the following tesla-suboptions:cc30
Generate code for compute capability 3.0.cc35
Generate code for compute capability 3.5.cc3x
Generate code for the lowest 3.x compute capability possible.cc3+
Is equivalent to cc3x.[no]debug
Enable[disable] debug information generation in device code.fastmath
Use routines from the fast math library.[no]flushz
Enable[disable] flush-to-zero mode for floating point computations in the GPUcode generated forPGI Accelerator model compute regions.
keepKeep the kernel files.
kepleris equivalent to cc3x.
kepler+is equivalent to cc3+.
llvmGenerate code using the llvm-based back-end.
[no]debugEnable[disable] GPU debug information generation.
deepcopyEnable full deep copy of aggregate data structions in OpenACC; Fortran only.
[no]lineinfoEnable[disable] GPU line information generation.
maxregcount:nSpecify the maximum number of registers to use on the GPU. Leaving this blankindicates no limit.
[no]fmaDo not generate fused multiply-add instructions.
noL1Prevents the use of L1 hardware data cache to cache global variables.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 50
pin+is equivalent to cc3+.
[no]rdcGenerate [do not generate] relocatable device code.
[no]requiredGenerate [do not generate] a compiler error if accelerator device code cannot begenerated.
teslais equivalent to -ta=tesla,cc2+
hostUse the host option to generate code to execute OpenACC regions on the host.
The -ta=host flag has no suboptions.multicore
Use the multicore option to generate OpenACC parallel regions to execute inparallel on individual host cores.
The -ta=multicore flag has no suboptions.
Multiple Targets
When host is one of the multiple targets, such as -ta=tesla,host, the result isgenerated code that can be run with or without an attached accelerator.
Relocatable Device Code
A rdc option is available for the -ta and -Mcuda flags that specifies to generaterelocatable device code. Starting in PGI 14.1, the default code generation and linkingmode for NVIDIA-target OpenACC and CUDA Fortran is rdc, relocatable device code.
You can disable the default and enable the old behavior and non-relocatable code byspecifying any of the following: -ta=tesla:nordc, -Mcuda=nordc.
LLVM and Native GPU Code Generation
For accelerator code generation, PGI 2017 has two options.
‣ The compilers generate an LLVM-based intermediate representation by default.‣ In legacy mode, the compilers generate low-level CUDA C code. To enable this code
generation, use -ta=tesla:nollvm on NVIDIA Tesla hardware.
DWARF Debugging Formats
PGI's debugging capability for Tesla uses the LLVM back-end. Use the compiler's-g option to enable the generation of full dwarf information on both the host anddevice; in the absence of other optimization flags, -g sets the optimization level tozero. If a -O option raises the optimization level to one or higher, only GPU line
Command-Line Options Reference
PVF Reference Guide Version 2017 | 51
information is generated on the device even when -g is specified. To enforce full dwarfgeneration for device code at optimization levels above zero, use the debug suboptionto -ta=tesla. Conversely, to prevent the generation of dwarf information for devicecode, use the nodebug suboption to -ta=tesla. Both debug and nodebug can beused independently of -g.
Related options
-#
2.2.49. -timePrint execution times for various compilation steps.
Default
The compiler does not print execution times for compilation steps.
Usage
In the following example, pgfortran prints the execution times for the variouscompilation steps.$ pgfortran -time myprog.f
Description
Use this option to print execution times for various compilation steps.
Related options
-#
2.2.50. -tp <target>[,target...]Sets the target processor.
Default
The PGI compilers produce code specifically targeted to the type of processor onwhich the compilation is performed. In particular, the default is to use all supportedinstructions wherever possible when compiling on a given system.
The default target processor is auto-selected depending on the processor on which thecompilation is performed. You can specify a target processor to compile for a differentprocessor type, such as to select a more generic processor, allowing the code to run onmore system types. Specifying two or more target processors enables unified binarycode generation, where two or more versions of each function may be generated, eachversion optimized for the specific instruction set available in each target processor.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 52
Executables created on a given system without the -tp flag may not be usable onprevious generation systems. For example, executables created on an Intel Sandybridgeprocessor may use instructions that are not available on earlier Intel Nehalem or Intel P7systems.
Syntax
Syntax for 64-bit targets:-tp {k8-64 | k8-64e | p7-64 | core2-64 | x64}
Usage
In the following example, pgfortran sets the target processor to a 64-bit Intel Nehalemprocessor:$ pgfortran -tp=nehalem-64 myprog.f
Description
Use this option to set the target architecture. By default, the PGI compiler uses allsupported instructions wherever possible when compiling on a given system.
Processor-specific optimizations can be specified or limited explicitly by using the -tpoption. Thus, it is possible to create executables that are usable on previous generationsystems.
To set this option in PVF, use the Fortran | Target Processors | Unified BinaryInformation property, described in ‘Unified Binary Information’.
The following list contains the possible suboptions for -tp and the processors that eachsuboption is intended to target. Options without a bit-length suffix use the current widthassociated with the driver on your path.barcelona
generate code for AMD Opteron/Quadcore and compatible processors.bulldozer
generate code for AMD Bulldozer and compatible processors.core2
generate code for Intel Core 2 Duo and compatible processors.haswell
generate code that is usable on any Haswell processor-based system.istanbul
generate code that is usable on any Istanbul processor-based system.k8
generate code hat is usable on any AMD64 and compatible processor.k8-64e
generate 64-bit code for AMD Opteron Revision E, AMD Turion, and compatibleprocessors.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 53
nehalemgenerate code that is usable on any Nehalem processor-based system.
p7generate code for Pentium 4 and compatible processors.
penryngenerate code for Intel Penryn Architecture and compatible processors.
piledrivergenerate code that is usable on any Piledriver processor-based system.
pxgenerate code that is usable on any x86-64 processor-based system.
sandybridgegenerate code for Intel Sandy Bridge and compatible processors.
shanghaigenerate code that is usable on any AMD Shanghai processor-based system.
x64generate 64-bit unified binary code including full optimizations and support for bothAMD and Intel x86-64 processors.
Refer to the PGI Release Notes for a concise list of the features of these processors thatdistinguish them as separate targets when using the PGI compilers and tools.
Using -tp to Generate a Unified Binary
Different processors have differences, some subtle, in hardware features such asinstruction sets and cache size. The compilers make architecture-specific decisions aboutsuch things as instruction selection, instruction scheduling, and vectorization. Anyof these decisions can have significant effects on performance and compatibility. PGIunified binaries provide a low-overhead means for a single program to run well on anumber of hardware platforms.
You can use the -tp option to produce PGI Unified Binary programs. The compilersgenerate, and combine into one executable, multiple binary code streams, eachoptimized for a specific platform. At runtime, this one executable senses theenvironment and dynamically selects the appropriate code stream.
The target processor switch, -tp , accepts a comma-separated list of 64-bit targets andwill generate code optimized for each listed target. For example, the following switchgenerates optimized code for three targets: k8-64, p7-64, and core2-64.
Syntax for optimizing for multiple targets:-tp k8-64,p7-64,core2-64
The -tp k8-64 and -tp k8-64e options result in generation of code supportedon and optimized for AMD x64 processors, while the -tp p7-64 option results ingeneration of code that is supported on and optimized for Intel x86-64 processors.Performance of k8-64 or k8-64e code executed on Intel x86-64 processors, or of p7-64
Command-Line Options Reference
PVF Reference Guide Version 2017 | 54
code executed on AMD x86-64 processors, can often be significantly less than thatobtained with a native binary.
The special -tp x64 option is equivalent to -tp k8-64,p7-64 . This switch producesPGI Unified Binary programs containing code streams fully optimized and supportedfor bothAMD64 and Intel 64 processors.
For more information on unified binaries, refer to the section ’Processor-SpecificOptimization and the Unified Binary’ in the PGI Compiler User’s Guide.
Related options
All -M<pgflag> options that control environments, as listed in Environment Controls
2.2.51. -[no]tracebackAdds debug information for runtime traceback for use with the environment variablePGI_TERM.
Default
The compiler enables traceback for FORTRAN and disables traceback for C and C++.
Syntax-traceback
Usage
In this example, pgfortran enables traceback for the program myprog.f.$ pgfortran -traceback myprog.f
Description
Use this option to enable or disable runtime traceback information for use with theenvironment variable PGI_TERM.
Setting setTRACEBACK=OFF; in siterc or .mypg*rc also disables default traceback.
Using ON instead of OFF enables default traceback.
Related options
None.
2.2.52. -uInitializes the symbol-table with <symbol>, which is undefined for the linker. Anundefined symbol triggers loading of the first member of an archive library.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 55
Default
The compiler does not use the -u option.
Syntax-usymbol
Where symbol is a symbolic name.
Usage
In this example, pgfortran initializes symbol-table with test.$ pgfortran -utest myprog.f
Description
Use this option to initialize the symbol-table with <symbol>, which is undefined for thelinker. An undefined symbol triggers loading of the first member of an archive library.
Related options
-c, -o
2.2.53. -UUndefines a preprocessor macro.
Syntax-Usymbol
Where symbol is a symbolic name.
Usage
The following examples undefine the macro test.
$ pgfortran -Utest myprog.F$ pgfortran -Dtest -Utest myprog.F
Description
Use this option to undefine a preprocessor macro. You can also use the #undef pre-processor directive to undefine macros.
To set this option in PVF, use the Fortran | Preprocessor | Undefine PreprocessorDefinitions property, described in ‘Undefine Preprocessor Definitions’.
Related options
-D, Mnostddef
Command-Line Options Reference
PVF Reference Guide Version 2017 | 56
2.2.54. -V[release_number]Displays additional information, including version messages. Further, if arelease_number is appended, the compiler driver attempts to compile using thespecified release instead of the default release.
There can be no space between -V and release_number.
Default
The compiler does not display version information and uses the release specified byyour path to compile.
Usage
The following command-line shows the output using the -V option.% pgfortran -V myprog.f
The following command-line causes pgcc to compile using the 5.2 release instead of thedefault release.% pgcc -V5.2 myprog.c
Description
Use this option to display additional information, including version messages or, if arelease_number is appended, to instruct the compiler driver to attempt to compile usingthe specified release instead of the default release.
The specified release must be co-installed with the default release, and must have arelease number greater than or equal to 4.1, which was the first release that supportedthis functionality.
To set this option in PVF, use the Fortran | General | Display Startup Banner property,described in ‘Display Startup Banner’.
Related options
-Minfo[=option [,option,...]], -v
2.2.55. -vDisplays the invocations of the compiler, assembler, and linker.
Default
The compiler does not display individual phase invocations.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 57
Usage
In the following example you use -v to see the commands sent to compiler tools,assembler, and linker.$ pgfortran -v myprog.f90
Description
Use the -v option to display the invocations of the compiler, assembler, and linker. Theseinvocations are command lines created by the compiler driver from the files and the -Woptions you specify on the compiler command-line.
Related options
-dryrun, -Minfo[=option [,option,...]], -V[release_number], -W
2.2.56. -WPasses arguments to a specific phase.
Syntax-W{0 | a | l },option[,option...]
You cannot have a space between the -W and the single-letter pass identifier,between the identifier and the comma, or between the comma and the option.
0(the number zero) specifies the compiler.
aspecifies the assembler.
l(lowercase letter l) specifies the linker.
optionis a string that is passed to and interpreted by the compiler, assembler or linker.Options separated by commas are passed as separate command line arguments.
Usage
In the following example the linker loads the text segment at address 0xffc00000 andthe data segment at address 0xffe00000.$ pgfortran -Wl,-k,-t,0xffc00000,-d,0xffe00000 myprog.f
Description
Use this option to pass arguments to a specific phase. You can use the -W option tospecify options for the assembler, compiler, or linker.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 58
A given PGI compiler command invokes the compiler driver, which parses thecommand-line, and generates the appropriate commands for the compiler, assembler,and linker.
Related options
-Minfo[=option [,option,...]], -V[release_number], -v
2.2.57. -wDo not print warning messages.
Default
The compiler prints warning messages.
Usage
In the following example no warning messages are printed.$ pgfortran -w myprog.f
Description
Use the -w option to not print warning messages. Sometimes the compiler issues manywarning in which you may have no interest. You can use this option to not issue thosewarnings.
Related options
-silent
2.3. -M Options by CategoryThis section describes each of the options available with -M by the categories:
Code Generation Fortran Language Controls Optimization Environment
C/C++ Language Controls Inlining Miscellaneous
The following sections provide detailed descriptions of several, but not all, of the-M<pgflag> options. For a complete alphabetical list of all the options, refer to Table10. These options are grouped according to categories and are listed with exact syntax,defaults, and notes concerning similar or related options.
2.3.1. Code Generation ControlsThis section describes the -M<pgflag> options that control code generation.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 59
Default: For arguments that you do not specify, the default code generation controls arethese:
nodaz norecursive nosecond_underscore
noflushz noreentrant nostride0
largeaddressaware noref_externals signextend
Related options: -D, -I, -L, -l, -U.
The following list provides the syntax for each -M<pgflag> option that controls codegeneration. Each option has a description and, if appropriate, any related options.-Mdaz
Set IEEE denormalized input values to zero; there is a performance benefit butmisleading results can occur, such as when dividing a small normalized number by adenormalized number. To take effect, this option must be set for the main program. To set this option in PVF,use the Fortran |
Floating Point Options | Treat Denormalized Values as Zero property, described in‘Treat Denormalized Values as Zero’
-MnodazDo not treat denormalized numbers as zero. To take effect, this option must be set for the main program.
-MnodwarfSpecifies not to add DWARF debug information. To take effect, this option must be used in combination with -g.
-Mdwarf1Generate DWARF1 format debug information. To take effect, this option must be used in combination with -g.
-Mdwarf2Generate DWARF2 format debug information. To take effect, this option must be used in combination with -g.
-Mdwarf3Generate DWARF3 format debug information. To take effect, this option must be used in combination with -g.
-MflushzSet SSE flush-to-zero mode; if a floating-point underflow occurs, the value is set tozero. To take effect, this option must be set for the main program.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 60
To set this option in PVF, use the Fortran | Floating Point Options | FlushDenormalized Results to Zero property, described in ‘Flush Denormalized Results toZero’ on page 391.
-MnoflushzDo not set SSE flush-to-zero mode; generate underflows. To take effect, this option must be set for the main program.
-Mfunc32Align functions on 32-byte boundaries.
-Minstrument[=functions] (linux86-64 only)Generate additional code to enable instrumentation of functions. The option-Minstrument=functions is the same as -Minstrument. Implies -Minfo=ccff and -Mframe.
-Mlargeaddressaware=[no][Win64 only] Generates code that allows for addresses greater than 2 GB, using RIP-relative addressing. Use-Mlargeaddressaware=no for a direct addressing mechanism that restricts thetotal addressable memory.
Do not use -Mlargeaddressaware=no if the object file will be placed in a DLL.
If -Mlargeaddressaware=no is used to compile any object file, it must also be usedwhen linking.
-Mlarge_arraysEnable support for 64-bit indexing and single static data objects larger than 2 GBin size. This option is the default in the presence of -mcmodel=medium. It can beused separately together with the default small memory model for certain 64-bitapplications that manage their own memory space. For more information, refer to the ‘Programming Considerations for 64-BitEnvironments’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
-Mnolarge_arraysDisable support for 64-bit indexing and single static data objects larger than 2 GB insize. When this option is placed after -mcmodel=medium on the command line, itdisables use of 64-bit indexing for applications that have no single data object largerthan 2 GB. For more information, refer to the ‘Programming Considerations for 64-BitEnvironments’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
-MnomainInstructs the compiler not to include the object file that calls the Fortran mainprogram as part of the link step. This option is useful for linking programs in which
Command-Line Options Reference
PVF Reference Guide Version 2017 | 61
the main program is written in C/C++ and one or more subroutines are written inFortran.
-Mmpi=option-Mmpi adds the include and library options to the compile and link commandsnecessary to build an MPI application using MPI header files and libraries. To use -Mmpi, you must have a version of MPI installed on your system. This option tells the compiler to use the headers and libraries for the specified versionof MPI.
-Mmpi=msmpi – Select the default Microsoft MPI libraries on Windows. For more information, refer to the ‘Programming Considerations for 64-BitEnvironments’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
-M[no]movntInstructs the compiler to generate nontemporal move and prefetch instructions evenin cases where the compiler cannot determine statically at compile-time that theseinstructions will be beneficial.
-M[no]preenables [disables] partial redundancy elimination.
-Mprof[=option[,option,...]]Set performance profiling options. Use of these options changes which sections areincluded in the binary. These sections can be read by the PGI profiler. The option argument can be any of the following:[no]ccff
Enable [disable] common compiler feedback format, CCFF, information.dwarf
Add limited DWARF symbol information sufficient for most performanceprofilers.
-Mrecursiveinstructs the compiler to allow Fortran subprograms to be called recursively.
-MnorecursiveFortran subprograms may not be called recursively.
-Mref_externalsforce references to names appearing in EXTERNAL statements.
-Mnoref_externalsdo not force references to names appearing in EXTERNAL statements.
-Mreentrantinstructs the compiler to avoid optimizations that can prevent code from beingreentrant.
-Mnoreentrantinstructs the compiler not to avoid optimizations that can prevent code from beingreentrant.
-Msecond_underscoreinstructs the compiler to add a second underscore to the name of a Fortran globalsymbol if its name already contains an underscore. This option is useful for
Command-Line Options Reference
PVF Reference Guide Version 2017 | 62
maintaining compatibility with object code compiled using g77, which uses thisconvention by default.
-Mnosecond_underscoreinstructs the compiler not to add a second underscore to the name of a Fortran globalsymbol if its name already contains an underscore.
-Msafe_lastvalWhen a scalar is used after a loop, but is not defined on every iteration of the loop,the compiler does not by default parallelize the loop. However, this option tells thecompiler it’s safe to parallelize the loop. For a given loop, the last value computed forall scalars makes it safe to parallelize the loop.
-Msignextendinstructs the compiler to extend the sign bit that is set as a result of converting anobject of one data type to an object of a larger signed data type.
-Mnosignextendinstructs the compiler not to extend the sign bit that is set as the result of convertingan object of one data type to an object of a larger data type.
-Mstack_arraysplaces automatic arrays on the stack.
-Mnostack_arraysallocates automatic arrays on the heap. -Mnostack_arrays is the default and whattraditionally has been the approach used.
-Mstride0instructs the compiler to inhibit certain optimizations and to allow for stride 0 arrayreferences. This option may degrade performance and should only be used if zero-stride induction variables are possible.
-Mnostride0instructs the compiler to perform certain optimizations and to disallow for stride 0array references.
-Mvarargsforce Fortran program units to assume procedure calls are to C functions with avarargs-type interface.
2.3.2. Environment ControlsThis section describes the -M<pgflag> options that control environments.
Default: For arguments that you do not specify, the default environment option dependson your configuration.
The following list provides the syntax for each -M<pgflag> option that controlsenvironments. Each option has a description and, if appropriate, a list of any relatedoptions.-Mnostartup
instructs the linker not to link in the standard startup routine that contains the entrypoint (_start) for the program.
If you use the -Mnostartup option and do not supply an entry point, the linkerissues the following error message: Warning: cannot find entry symbol _start
Command-Line Options Reference
PVF Reference Guide Version 2017 | 63
-M[no]smartalloc[=huge|huge:<n>|hugebss|nohuge]adds a call to the routine mallopt in the main routine. This option supports largeTLBs on Linux and Windows. This option must be used to compile the main routineto enable optimized malloc routines. The option arguments can be any of the following:huge
Link in the huge page runtime library. Enables large 2-megabyte pages to be allocated. The effect is to reduce the numberof TLB entries required to execute a program. This option is most effective onBarcelona and Core 2 systems; older architectures do not have enough TLB entriesfor this option to be beneficial. By itself, the huge suboption tries to allocate asmany huge pages as required.
huge:<n>Link the huge page runtime library and allocate n huge pages. Use this suboptionto limit the number of huge pages allocated to n. You can also limit the pages allocated by using the environment variablePGI_HUGE_PAGES.
hugebss(64-bit only) Puts the BSS section in huge pages; attempts to put a program'suninitialized data section into huge pages.
This flag dynamically links the library libhugetlbfs_pgi even if-Bstatic is used.
nohugeOverrides a previous -Msmartalloc=huge setting.
Tip To be effective, this switch must be specified when compiling the filecontaining the Fortran, C, or C++ main program.
-Mnostdincinstructs the compiler to not search the standard location for include files. To setthis option in PVF, use the Fortran | Preprocessor | Ignore Standard Include Pathproperty, described in ‘Ignore Standard Include Path’ on page 381.
-Mnostdlibinstructs the linker not to link in the standard libraries in the library directory libwithin the standard directory. You can link in your own library with the -l option orspecify a library directory with the -L option.
2.3.3. Fortran Language ControlsThis section describes the -M<pgflag> options that affect Fortran languageinterpretations by the PGI Fortran compilers. These options are valid only for theFortran compiler drivers.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 64
Default: Before looking at all the options, let's look at the defaults. For arguments thatyou do not specify, the defaults are as follows:
backslash nodefaultunit dollar,_ noonetrip nounixlogical
nodclchk nodlines noiomutex nosave noupcase
The following list provides the syntax for each -M<pgflag> option that affect Fortranlanguage interpretations. Each option has a description and, if appropriate, a list of anyrelated options.-Mallocatable=95|03
controls whether Fortran 95 or Fortran 2003 semantics are used in allocatable arrayassignments. The default behavior is to use Fortran 95 semantics; the 03 optioninstructs the compiler to use Fortran 2003 semantics.
-Mbackslashinstructs the compiler to treat the backslash as a normal character, and not as anescape character in quoted strings.
-Mnobackslashinstructs the compiler to recognize a backslash as an escape character in quotedstrings (in accordance with standard C usage).
-Mcudainstructs the compiler to enable CUDA Fortran.
The following suboptions exist:
If more than one option is on the command line, all the specified options occur.
cc30Generate code for compute capability 3.0.
cc35Generate code for compute capability 3.5.
cc3xGenerate code for the lowest 3.x compute capability possible.
cc3+Is equivalent to cc3x.
cc50Generate code for compute capability 5.0.
cc60Generate code for compute capability 6.0.
cuda7.5 or 7.5Specify the NVIDIA CUDA 7.5 version of the toolkit. This is the default.
cuda8.0 or 8.0Specify the NVIDIA CUDA 8.0 version of the toolkit.
Compile with the CUDA 7.5 or CUDA 8.0 toolkit either by using the -Mcuda=7.5or -Mcuda=8.0 option, or by adding set DEFCUDAVERSION=7.5 or setDEFCUDAVERSION=8.0 to the siterc file. This action generates binaries thatmay not work on machines with an earlier CUDA driver.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 65
pgaccelinfo prints the driver version as the first line of output.
For a 7.5 driver: CUDA Driver Version 7050For an 8.0 driver: CUDA Driver Version 8000
emuEnable CUDA Fortran emulation mode.
fastmathUse routines from the fast math library.
fermiis equivalent to -Mcuda,cc2x
[no]flushzEnable[disable] flush-to-zero mode for floating point computations in the GPUcode generated for CUDA Fortran kernels.
generate rdcGenerate relocatable device code
keepbinKeep the generated binary (.bin) file for CUDA Fortran.
keepgpuKeep the generated GPU code for CUDA Fortran.
keepptxKeep the portable assembly (.ptx) file for the GPU code.
kepleris equivalent to -Mcuda,cc3x
llvmGenerate code using the llvm-based back-end.
[no]debugEnable[disable] GPU debug information generation.
[no]lineinfoEnable[disable] GPU line information generation.
maxregcount:nSpecify the maximum number of registers to use on the GPU. Leaving this blankindicates no limit.
nofmaDo not generate fused multiply-add instructions.
noL1Prevent the use of L1 hardware data cache to cache global variables.
ptxinfoShow PTXAS informational messages during compilation.
rdcEnable CUDA Fortran separate compilation and linking of device routines,including device routines in Fortran modules. To enable separate compilation and linking, include the command line option -Mcuda=rdc on both the compile and the link steps.
-Mdclchkinstructs the compiler to require that all program variables be declared.
-Mnodclchkinstructs the compiler not to require that all program variables be declared.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 66
-Mdefaultunitinstructs the compiler to treat "*" as a synonym for standard input for reading andstandard output for writing.
-Mnodefaultunitinstructs the compiler to treat "*" as a synonym for unit 5 on input and unit 6 onoutput.
-Mdlinesinstructs the compiler to treat lines containing "D" in column 1 as executablestatements (ignoring the "D").
-Mnodlinesinstructs the compiler not to treat lines containing "D" in column 1 as executablestatements. The compiler does not ignore the "D".
-Mdollar,charchar specifies the character to which the compiler maps the dollar sign. The compilerallows the dollar sign in names.
-Mextendinstructs the compiler to accept 132-column source code; otherwise it accepts 72-column code.
-Mfixedinstructs the compiler to assume input source files are in FORTRAN 77-style fixedform format.
-Mfreeinstructs the compiler to assume input source files are in Fortran 90/95 freeformformat.
-Miomutexinstructs the compiler to generate critical section calls around Fortran I/O statements.
-Mnoiomutexinstructs the compiler not to generate critical section calls around Fortran I/Ostatements.
-Monetripinstructs the compiler to force each DO loop to execute at least once. This option isuseful for programs written for earlier versions of Fortran.
-Mnoonetripinstructs the compiler not to force each DO loop to execute at least once.
-Msaveinstructs the compiler to assume that all local variables are subject to the SAVEstatement. This may allow older Fortran programs to run, but it can greatly reduce performance.
-Mnosaveinstructs the compiler not to assume that all local variables are subject to the SAVEstatement.
-Mstandardinstructs the compiler to flag non-ANSI-conforming source code.
-Munixlogicaldirects the compiler to treat logical values as true if the value is non-zero and falseif the value is zero (UNIX F77 convention). When -Munixlogical is enabled, a logicalvalue or test that is non-zero is .TRUE., and a value or test that is zero is .FALSE.. In
Command-Line Options Reference
PVF Reference Guide Version 2017 | 67
addition, the value of a logical expression is guaranteed to be one (1) when the resultis .TRUE..
-Mnounixlogicaldirects the compiler to use the VMS convention for logical values for true and false.Even values are true and odd values are false.
-Mupcaseinstructs the compiler to preserve uppercase letters in identifiers. With -Mupcase, the identifiers "X" and "x" are different. Keywords must be in lowercase. This selection affects the linking process. If you compile and link the same sourcecode using -Mupcase on one occasion and -Mnoupcase on another, you may get twodifferent executables – depending on whether the source contains uppercase letters.The standard libraries are compiled using the default -Mnoupcase .
-Mnoupcaseinstructs the compiler to convert all identifiers to lower case. This selection affects the linking process. If you compile and link the same sourcecode using -Mupcase on one occasion and -Mnoupcase on another, you may get twodifferent executables, depending on whether the source contains uppercase letters.The standard libraries are compiled using -Mnoupcase.
2.3.4. Inlining ControlsThis section describes the -M<pgflag> options that control function inlining.
Usage:Before looking at all the options, let’s look at a couple examples. In the followingexample, the compiler extracts functions that have 500 or fewer statements from thesource file myprog.f and saves them in the file extract.il.$ pgfortran -Mextract=500 -o extract.il myprog.f
In the following example, the compiler inlines functions with fewer than approximately100 statements in the source file myprog.f.$ pgfortran -Minline=maxsize:100 myprog.f
Related options: -o, -Mextract
The following list provides the syntax for each -M<pgflag> option that controls functioninlining. Each option has a description and, if appropriate, a list of any related options.- M[no]autoinline[=option[,option,...]]
instructs the compiler to inline [not to inline] a C/C++ function at -O2, where theoption can be any of these:maxsize:n
instructs the compiler not to inline functions of size > n. The default size is 100.totalsize:n
instructs the compiler to stop inlining when the size equals n. The default size is800.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 68
-Mextract[=option[,option,...]]Extracts functions from the file indicated on the command line and creates orappends to the specified extract directory where option can be any of the following:name:func
instructs the extractor to extract function func from the file.size:number
instructs the extractor to extract functions with number or fewer statements fromthe file.
lib:filename.extinstructs the extractor to use directory filename.ext as the extract directory,which is required to save and re-use inline libraries.
If you specify both name and size, the compiler extracts functions that match func,or that have number or fewer statements. For examples of extracting functions,refer to the ‘Using Function Inlining’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
-Minline[=option[,option,...]]instructs the compiler to pass options to the function inliner, where the option can beany of the following:except:func
Inlines all eligible functions except func, a function in the source text. You can usea comma-separated list to specify multiple functions.
[name:]funcInlines all functions in the source text whose name matches func. You can use acomma-separated list to specify multiple functions.
The function name should be a non-numeric string that does not contain a period.You can also use a name: prefix followed by the function name. If name: isspecified, what follows is always the name of a function.
[maxsize:]numberA numeric option is assumed to be a size. Functions of size number or less areinlined. If both number and function are specified, then functions matching thegiven name(s) or meeting the size requirements are inlined.
The size number need not exactly equal the number of statements in a selectedfunction; the size parameter is merely a rough guage.
[no]reshapeinstructs the inliner to allow [disallow] inlining in Fortran even when array shapesdo not match. The default is -Minline=noreshape, except with -Mconcur or-mp, where the default is -Minline=reshape,=reshape.
smallsize:numberAlways inline functions of size smaller than number regardless of other size limits.
totalsize:numberStop inlining in a function when the function's total inlined size reaches thenumber specified.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 69
[lib:]filename.extinstructs the inliner to inline the functions within the library file filename.ext.The compiler assumes that a filename.ext option containing a period is alibrary file.
Tip Create the library file using the -Mextract option. You can also use alib: prefix followed by the library name.
‣ If lib: is specified, no period is necessary in the library name. Functionsfrom the specified library are inlined.
‣ If no library is specified, functions are extracted from a temporary librarycreated during an extract prepass.
If you specify both func and number, the compiler inlines functions that match thefunction name or have number or fewer statements.
Inlining can be disabled with -Mnoinline.
To set this option in PVF, use the Fortran | Optimization | Inlining property,described in ‘Inlining’
For examples of inlining functions, refer to ‘Using Function Inlining’ in the PGICompiler User’s Guide.
2.3.5. Optimization ControlsThis section describes the -M<pgflag> options that control optimization.
Default: Before looking at all the options, let's look at the defaults. For arguments thatyou do not specify, the default optimization control options are as follows:
depchk noipa nounroll nor8
i4 nolre novect nor8intrinsics
nofprelaxed noprefetch
If you do not supply an option to -Mvect, the compiler uses defaults that aredependent upon the target system.
Usage: In this example, the compiler invokes the vectorizer with use of packed SSEinstructions enabled.>$ pgfortran -Mvect=sse -Mcache_align myprog.f
Related options: -g, -O
The following list provides the syntax for each -M<pgflag> option that controlsoptimization. Each option has a description and, if appropriate, a list of any relatedoptions.-Mcache_align
Align unconstrained objects of length greater than or equal to 16 bytes on cache-line boundaries. An unconstrained object is a data object that is not a member of an
Command-Line Options Reference
PVF Reference Guide Version 2017 | 70
aggregate structure or common block. This option does not affect the alignment ofallocatable or automatic arrays. To effect cache-line alignment of stack-based local variables, the main program orfunction must be compiled with -Mcache_align.
-Mconcur[=option [,option,...]]Instructs the compiler to enable auto-concurrentization of loops. If -Mconcur isspecified, multiple processors will be used to execute loops that the compilerdetermines to be parallelizable. option is one of the following:allcores
Instructs the compiler to use all available cores. Use this option at link time.[no]altcode:n
Instructs the parallelizer to generate alternate serial code for parallelized loops.
‣ If altcode is specified without arguments, the parallelizer determines anappropriate cutoff length and generates serial code to be executed wheneverthe loop count is less than or equal to that length.
‣ If altcode:n is specified, the serial altcode is executed whenever the loop countis less than or equal to n.
‣ If noaltcode is specified, the parallelized version of the loop is always executedregardless of the loop count.
cncallIndicates that calls in parallel loops are safe to parallelize. Loops containing calls are candidates for parallelization. Also, no minimum loopcount threshold must be satisfied before parallelization will occur, and last valuesof scalars are assumed to be safe.
[no]innermostInstructs the parallelizer to enable parallelization of innermost loops. The defaultis to not parallelize innermost loops, since it is usually not profitable on dual-coreprocessors.
noassocInstructs the parallelizer to disable parallelization of loops with reductions.
When linking, the -Mconcur switch must be specified or unresolved references result.The NCPUS environment variable controls how many processors or cores are used toexecute parallelized loops.
To set this option in PVF, use the Fortran | Optimization | Auto-Parallelizationproperty, described in ‘Auto-Parallelization’.
This option applies only on shared-memory multi-processor (SMP) or multicoreprocessor-based systems.
-Mcray[=option[,option,...]]Force Cray Fortran (CF77) compatibility with respect to the listed options. Possiblevalues of option include:
Command-Line Options Reference
PVF Reference Guide Version 2017 | 71
pointerfor purposes of optimization, it is assumed that pointer-based variables do notoverlay the storage of any other variable.
-Mdepchkinstructs the compiler to assume unresolved data dependencies actually conflict.
-MnodepchkInstructs the compiler to assume potential data dependencies do not conflict.However, if data dependencies exist, this option can produce incorrect code.
-MdseEnables a dead store elimination phase that is useful for programs that rely onextensive use of inline function calls for performance. This is disabled by default.
-MnodseDisables the dead store elimination phase. This is the default.
-M[no]fpapprox[=option]Perform certain floating point operations using low-precision approximation. -Mnofpapprox specifies not to use low-precision fp approximation operations. By default -Mfpapprox is not used. If -Mfpapprox is used without suboptions, it defaults to use approximate div, sqrt,and rsqrt. The available suboptions are these:div
Approximate floating point divisionsqrt
Approximate floating point square rootrsqrt
Approximate floating point reciprocal square root-M[no]fpmisalign
Instructs the compiler to allow (not allow) vector arithmetic instructions withmemory operands that are not aligned on 16-byte boundaries. The default is-Mnofpmisalign on all processors.
Applicable only with one of these options: -tp barcelona or -tp barcelona-64 ornewer processors.
-M[no]fprelaxed[=option]Instructs the compiler to use [not use] relaxed precision in the calculation of someintrinsic functions. Can result in improved performance at the expense of numericalaccuracy.
To set this option in PVF, use the Fortran | Floating Point Options | Floating PointConsistency property. For more information on this property, refer to ‘Floating PointConsistency’. The possible values for option are:div
Perform divide using relaxed precision.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 72
intrinsicEnables use of relaxed precision intrinsics.
noorderDo not allow expression reordering or factoring.
orderAllow expression reordering, including factoring.
recipPerform reciprocal using relaxed precision.
rsqrtPerform reciprocal square root (1/sqrt) using relaxed precision.
sqrtPerform square root with relaxed precision.
With no options, -Mfprelaxed generates relaxed precision code for those operationsthat generate a significant performance improvement, depending on the targetprocessor. The default is -Mnofprelaxed which instructs the compiler to not use relaxedprecision in the calculation of intrinsic functions.
-Mi4instructs the compiler to treat INTEGER variables as INTEGER*4.
-Mlre[=array | assoc | noassoc]Enables loop-carried redundancy elimination, an optimization that can reduce thenumber of arithmetic operations and memory references in loops. The availablesuboptions are:array
treat individual array element references as candidates for possible loop-carriedredundancy elimination. The default is to eliminate only redundant expressionsinvolving two or more operands.
assocallow expression re-association. Specifying this suboption can increaseopportunities for loop-carried redundancy elimination but may alter numericalresults.
noassocdisallow expression re-association.
-MnolreDisable loop-carried redundancy elimination.
-MnoframeEliminate operations that set up a true stack frame pointer for every function. Withthis option enabled, you cannot perform a traceback on the generated code and youcannot access local variables.
To set this option in PVF, use the Fortran | Optimization | Use Frame Pointerproperty, described in ‘Use Frame Pointer’
-Mnoi4instructs the compiler to treat INTEGER variables as INTEGER*2.
-MpreEnables partial redundancy elimination.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 73
-Mprefetch[=option [,option...]]enables generation of prefetch instructions on processors where they are supported.Possible values for option include:d:m
set the fetch-ahead distance for prefetch instructions to m cache lines.n:p
set the maximum number of prefetch instructions to generate for a given loop to p.nta
use the prefetch instruction.plain
use the prefetch instruction (default).t0
use the prefetcht0 instruction.w
use the AMD-specific prefetchw instruction.-Mnoprefetch
Disables generation of prefetch instructions.-M[no]propcond
Enables or disables constant propagation from assertions derived from equalityconditionals. The default is enabled.
-Mr8The compiler promotes REAL variables and constants to DOUBLE PRECISIONvariables and constants, respectively. DOUBLE PRECISION elements are 8 bytes inlength.
-Mnor8The compiler does not promote REAL variables and constants to DOUBLEPRECISION. REAL variables will be single precision (4 bytes in length).
-Mr8intrinsicsThe compiler treats the intrinsics CMPLX and REAL as DCMPLX and DBLE, respectively.
-Mnor8intrinsicsThe compiler does not promote the intrinsics CMPLX and REAL to DCMPLX and DBLE,respectively.
-MscalarsseUse SSE/SSE2 instructions to perform scalar floating-point arithmetic. This option isvalid only on option -tp [p7 | k8-32 | k8-64] targets.
-MnoscalarsseDo not use SSE/SSE2 instructions to perform scalar floating-point arithmetic; use x87instructions instead. This option is not valid in combination with the -tp k8-64 option.
-Msmartinstructs the compiler driver to invoke a post-pass assembly optimization utility.
-Mnosmartinstructs the compiler not to invoke an AMD64-specific post-pass assemblyoptimization utility.
-Munroll[=option [,option...]]invokes the loop unroller to execute multiple instances of the loop during eachiteration. This also sets the optimization level to 2 if the level is set to less than 2, or ifno -O or -g options are supplied. The option is one of the following:
Command-Line Options Reference
PVF Reference Guide Version 2017 | 74
c:minstructs the compiler to completely unroll loops with a constant loop count lessthan or equal to m, a supplied constant. If this value is not supplied, the m count isset to 4.
m:<n>instructs the compiler to unroll multi-block loops n times. This option is useful forloops that have conditional statements. If n is not supplied, then the default valueis 4. The default setting is not to enable -Munroll=m.
n:<n>instructs the compiler to unroll single-block loops n times, a loop that is notcompletely unrolled, or has a non-constant loop count. If n is not supplied, theunroller computes the number of times a candidate loop is unrolled.
To set this option in PVF, use the Fortran | Optimization | Loop Unroll Countproperty, described in ‘Loop Unroll Count’
-Mnounrollinstructs the compiler not to unroll loops.
-M[no]vect[=option [,option,...]]enable [disable] the code vectorizer, where option is one of the following:altcode
Instructs the vectorizer to generate alternate code (altcode) for vectorized loopswhen appropriate. For each vectorized loop the compiler decides whether togenerate altcode and what type or types to generate, which may be any or all of:altcode without iteration peeling, altcode with non-temporal stores and otherdata cache optimizations, and altcode based on array alignments calculateddynamically at runtime. The compiler also determines suitable loop count andarray alignment conditionals for executing the altcode. This option is enabled bydefault.
noaltcodeInstructs the vectorizer to disable alternate code generation for vectorized loops.
assocInstructs the vectorizer to enable certain associativity conversions that can changethe results of a computation due to roundoff error. A typical optimization is tochange an arithmetic operation to an arithmetic operation that is mathematicallycorrect, but can be computationally different, due to round-off error.
noassocInstructs the vectorizer to disable associativity conversions.
cachesize:nInstructs the vectorizer, when performing cache tiling optimizations, to assume acache size of n. The default is set per processor type, either using the -tp switch orauto-detected from the host computer.
[no]gatherInstructs the vectorizer to vectorize loops containing indirect array references, suchas this one:sum = 0.d0do k=d(j),d(j+1)-1 sum = sum + a(k)*b(c(k))enddo
The default is gather.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 75
partialInstructs the vectorizer to enable partial loop vectorization through innermost loopdistribution.
prefetchInstructs the vectorizer to search for vectorizable loops and, wherever possible,make use of prefetch instructions.
[no]shortInstructs the vectorizer to enable [disable] short vector operations. -Mvect=shortenables generation of packed SIMD instructions for short vector operations thatarise from scalar code outside of loops or within the body of a loop iteration.
[no]sizelimitInstructs the vectorizer to generate vector code for all loops where possibleregardless of the number of statements in the loop. This overrides a heuristic inthe vectorizer that ordinarily prevents vectorization of loops with a number ofstatements that exceeds a certain threshold. The default is nosizelimit.
smallvect[:n]Instructs the vectorizer to assume that the maximum vector length is less thanor equal to n. The vectorizer uses this information to eliminate generation of thestripmine loop for vectorized loops wherever possible. If the size n is omitted, thedefault is 100.
No space is allowed on either side of the colon (:).
[no]sseInstructs the vectorizer to search for vectorizable loops and, wherever possible,make use of SSE, SSE2, and prefetch instructions. The default is nosse.
[no]uniformInstructs the vectorizer to perform the same optimizations in the vectorized andresidual loops.
This option may affect the performance of the residual loop.
To set this option in PVF, use the Fortran | Optimization Vectorization property,described in ‘Vectorization’
-Mnovectinstructs the compiler not to perform vectorization. You can use this option tooverride a previous instance of -Mvect on the command-line, in particular for cases inwhich -Mvect is included in an aggregate option such as -fastsse.
-Mvect=[option]instructs the compiler to enable loop vectorization, where option is one of thefollowing:partial
Enable partial loop vectorization through innermost loop distribution.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 76
[no]shortEnable [disable] short vector operations. Enables [disables] generation of packedSIMD instructions for short vector operations that arise from scalar code outside ofloops or within the body of a loop iteration.
simd[:{128|256}]Specifies to vectorize using SIMD instructions and data, either 128 bits or 256 bitswide, on processors where there is a choice.
tileEnable tiling/blocking over multiple nested loops for more efficient cacheutilization.
-Mnovintrinstructs the compiler not to perform idiom recognition or introduce calls to hand-optimized vector functions.
2.3.6. Miscellaneous ControlsThis section describes the -M<pgflag> options that do not easily fit into one of the othercategories of -M<pgflag> options.
Default: Before looking at all the options, let’s look at the defaults. For arguments thatyou do not specify, the default miscellaneous options are as follows:
inform nobounds nolist warn
Related options: -m, -S, -V, -v
Usage: In the following example, the compiler includes Fortran source code with theassembly code. $ pgfortran -Manno -S myprog.f
In the following example, the assembler does not delete the assembly file myprog.safter the assembly pass. $ pgfortran -Mkeepasm myprog.f
In the following example, the compiler displays information about inlined functionswith fewer than approximately 20 source lines in the source file myprog.f. $ pgfortran -Minfo=inline -Minline=20 myprog.f
In the following example, the compiler creates the listing file myprog.lst. $ pgfortran -Mlist myprog.f
In the following example, array bounds checking is enabled. $ pgfortran -Mbounds myprog.f
The following list provides the syntax for each miscellaneous -M<pgflag> option. Eachoption has a description and, if appropriate, a list of any related options.-Manno
annotate the generated assembly code with source code. Implies -Mkeepasm.
To set this option in PVF, use the Fortran | Output | Annotated ASM Listingproperty, described in ‘Annotate Assembly’.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 77
-Mboundsenables array bounds checking.
‣ If an array is an assumed size array, the bounds checking only applies to thelower bound.
‣ If an array bounds violation occurs during execution, an error message describingthe error is printed and the program terminates. The text of the error messageincludes the name of the array, the location where the error occurred (the sourcefile and the line number in the source), and information about the out of boundssubscript (its value, its lower and upper bounds, and its dimension).
The following is a sample error message:PGFTN-F-Subscript out of range for array a (a.f: 2) subscript=3, lower bound=1, upper bound=2, dimension=2
-Mnoboundsdisables array bounds checking.
-Mbyteswapioswap byte-order from big-endian to little-endian or vice versa upon input/output ofFortran unformatted data files.
-Mchkptrinstructs the compiler to check for pointers that are dereferenced while initialized toNULL.
-Mchkstkinstructs the compiler to check the stack for available space in the prologue of afunction and before the start of a parallel region. Prints a warning message and abortsthe program gracefully if stack space is insufficient. This option is useful when many local and private variables are declared in anOpenMP program. If the user also sets the PGI_STACK_USAGE environment variable to any value, thenthe program displays the stack space allocated and used after the program exits. Forexample, you might see something similar to the following message:thread 0 stack: max 8180KB, used 48KB
This message indicates that the program used 48KB of a 8180KB allocated stack. Thisinformation is useful when you want to explicitly set a reserved and committed stacksize for your programs, such as using the -stack option on Windows. For more information on the PGI_STACK_USAGE, refer to ‘PGI_STACK_USAGE’ inthe PGI Compiler User’s Guide.
-Mcpp[=option [,option,...]]run the PGI cpp-like preprocessor without execution of any subsequent compilationsteps. This option is useful for generating dependence information to be included inmakefiles.
Only one of the m, md, mm or mmd options can be present; if multiple of theseoptions are listed, the last one listed is accepted and the others are ignored.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 78
The option is one or more of the following:m
print makefile dependencies to stdout.md
print makefile dependencies to filename.d, where filename is the root name ofthe input file being processed, ignoring system include files.
mmprint makefile dependencies to stdout, ignoring system include files.
mmdprint makefile dependencies to filename.d, where filename is the root name ofthe input file being processed, ignoring system include files.
[no]commentdo [do not] retain comments in output.
[suffix:]<suff>use <suff> as the suffix of the output file containing makefile dependencies.
-MdllThis Windows-only flag has been deprecated. Refer to -Bdynamic. This flag wasused to link with the DLL versions of the runtime libraries, and it was required whenlinking with any DLL built by any PGI compilers. This option implied -D_DLL,which defines the preprocessor symbol _DLL.
-Mgccbug[s]instructs the compiler to match the behavior of certain gcc bugs.
-Miface[=option]adjusts the calling conventions for Fortran, where option is one of the following:cref
uses CREF calling conventions, no trailing underscores.mixed_str_len_arg
places the lengths of character arguments immediately after their correspondingargument. Has affect only with the CREF calling convention.
nomixed_str_len_argplaces the lengths of character arguments at the end of the argument list. Hasaffect only with the CREF calling convention.
-Minfo[=option [,option,...]]instructs the compiler to produce information on standard error, where option is oneof the following:all
instructs the compiler to produce all available -Minfo information. Implies anumber of suboptions:-Mneginfo=accel,inline,ipa,loop,lre,mp,opt,par,vect
accelinstructs the compiler to enable accelerator information.
ccffinstructs the compiler to append common compiler feedback format information,such as optimization information, to the object file.
ftninstructs the compiler to enable Fortran-specific information.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 79
inlineinstructs the compiler to display information about extracted or inlined functions.This option is not useful without either the -Mextract or -Minline option.
intensityinstructs the compiler to provide informational messages about the intensity of theloop. Specify <n> to get messages on nested loops.
‣ For floating point loops, intensity is defined as the number of floating pointoperations divided by the number of floating point loads and stores.
‣ For integer loops, the loop intensity is defined as the total number of integerarithmetic operations, which may include updates of loop counts andaddresses, divided by the total number of integer loads and stores.
‣ By default, the messages just apply to innermost loops.
ipainstructs the compiler to display information about interprocedural optimizations.
loopinstructs the compiler to display information about loops, such as information onvectorization.
lreinstructs the compiler to enable LRE, loop-carried redundancy elimination,information.
mpinstructs the compiler to display information about parallelization.
optinstructs the compiler to display information about optimization.
parinstructs the compiler to enable parallelizer information.
pfoinstructs the compiler to enable profile feedback information.
timeinstructs the compiler to display compilation statistics.
unrollinstructs the compiler to display information about loop unrolling.
vectinstructs the compiler to enable vectorizer information.
-Minform=levelinstructs the compiler to display error messages at the specified and higher levels,where level is one of the following:fatal
instructs the compiler to display fatal error messages.[no]file
instructs the compiler to print or not print source file names as they are compiled.The default is to print the names: -Minform=file.
informinstructs the compiler to display all error messages (inform, warn, severe andfatal).
severeinstructs the compiler to display severe and fatal error messages.
Command-Line Options Reference
PVF Reference Guide Version 2017 | 80
warninstructs the compiler to display warning, severe and fatal error messages.
To set this option in PVF, use the Fortran | Diagnostics | Warning Level property,described in ‘Warning Level’.
-Minstrumentation=optionspecifies the level of instrumentation calls generated. This option implies -Minfo=ccff, -Mframe. option is one of the following:level
specifies the level of instrumentation calls generated.function (default)
generates instrumentation calls for entry and exit to functions. Just after function entry and just before function exit, the following profilingfunctions are called with the address of the current function and its call site.(linux86-64 only).
void __cyg_profile_func_enter (void *this_fn, void *call_site);void __cyg_profile_func_exit (void *this_fn, void *call_site);
In these calls, the first argument is the address of the start of the current function.
To set this option in PVF, use the Fortran | Diagnostics | Warning Level property,described in ‘Warning Level’.
-Mkeepasminstructs the compiler to keep the assembly file as compilation continues. Normally,the assembler deletes this file when it is finished. The assembly file has the samefilename as the source file, but with a .s extension.
To set this option in PVF, use the Fortran | Output | Assembler Output property,described in ‘Generate Assembly’.
-Mlistinstructs the compiler to create a listing file. The listing file is filename.lst, wherethe name of the source file is filename.f.
-Mmakedllgenerate a dynamic link library (DLL).
-Mmakeimplibgenerate an import library for a DLL without creating the DLL. When used without -def:deffile, passes the switch -def to the librarian without a deffile.
-Mnames=lowercase|uppercasespecifies the case for the names of Fortran externals.
‣ lowercase - Use lowercase for Fortran externals.‣ uppercase - Use uppercase for Fortran externals.
-Mneginfo[=option [,option,...]]instructs the compiler to produce information on standard error, where option is oneof the following:
Command-Line Options Reference
PVF Reference Guide Version 2017 | 81
allinstructs the compiler to produce all available information on why variousoptimizations are not performed.
accelinstructs the compiler to enable accelerator information.
ccffinstructs the compiler to append information, such as optimization information, tothe object file.
concurinstructs the compiler to produce all available information on why loops arenot automatically parallelized. In particular, if a loop is not parallelized due topotential data dependence, the variable(s) that cause the potential dependence arelisted in the messages that you see when using the option -Mneginfo.
ftninstructs the compiler to enable Fortran-specific information.
inlineinstructs the compiler to display information about extracted or inlined functions.This option is not useful without either the -Mextract or -Minline option.
ipainstructs the compiler to display information about interprocedural optimizations.
loopinstructs the compiler to display information about loops, such as information onvectorization.
lreinstructs the compiler to enable LRE, loop-carried redundancy elimination,information.
mpinstructs the compiler to display information about parallelization.
optinstructs the compiler to display information about optimization.
parinstructs the compiler to enable parallelizer information.
pfoinstructs the compiler to enable profile feedback information.
vectinstructs the compiler to enable vectorizer information.
-Mnolistthe compiler does not create a listing file. This is the default.
-Mnoopenmpwhen used in combination with the -mp option, the compiler ignores OpenMPparallelization directives or pragmas, but still processes SGI-style parallelizationdirectives or pragmas.
-Mnosgimpwhen used in combination with the -mp option, the compiler ignores SGI-styleparallelization directives, but still processes OpenMP parallelization directives orpragmas.
-Mnopgdllmain(Windows only) do not link the module containing the default DllMain() into theDLL. This flag applies to building DLLs with the PGFORTRAN compilers. If you
Command-Line Options Reference
PVF Reference Guide Version 2017 | 82
want to replace the default DllMain() routine with a custom DllMain(), use this flagand add the object containing the custom DllMain() to the link line. The latest versionof the default DllMain() used by PGFORTRAN is included in the Release Notes foreach release. The PGFORTRAN-specific code in this routine must be incorporatedinto the custom version of DllMain() to ensure the appropriate function of your DLL.
-Mpreprocessinstruct the compiler to perform cpp-like preprocessing on assembly and Fortraninput source files.
To set this option in PVF, use the Fortran | Preprocessor | Preprocess Source Fileproperty, described in ‘Preprocessor Definitions’.
-Mwritable_stringsstores string constants in the writable data segment.
Options -Xs and -Xst include -Mwritable_strings.
PVF Reference Guide Version 2017 | 83
Chapter 3.DIRECTIVES REFERENCE
PGI Fortran compilers support proprietary directives. These directives overridecorresponding command-line options. For usage information such as the scope andrelated command-line options, refer to the PGI Compiler User’s Guide.
This section contains detailed descriptions of PGI’s proprietary directives.
3.1. PGI Proprietary Fortran Directive SummaryDirectives (Fortran comments) may be supplied by the user in a source file to provideinformation to the compiler. Directives alter the effects of certain command line optionsor default behavior of the compiler. They provide pragmatic information that controlthe actions of the compiler in a particular portion of a program without affecting theprogram as a whole. That is, while a command line option affects the entire source filethat is being compiled, directives apply, or disable, the effects of a command line optionto selected subprograms or to selected loops in the source file, for example, to optimize aspecific area of code. Use directives to tune selected routines or loops.
The Fortran directives may have any of the following forms:!pgi$g directive!pgi$r directive!pgi$l directive!pgi$ directive
where the scope indicator follows the $ and is either g (global), r (routine), or l (loop).This indicator controls the scope of the directive, though some directives ignore thescope indicator.
If the input is in fixed format, the comment character, !, * or C, must begin in column1.
Directives override corresponding command-line options. For usage information suchas the scope and related command-line options, refer to the the ‘Using Directives andPragmas’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
Directives Reference
PVF Reference Guide Version 2017 | 84
3.1.1. altcode (noaltcode)The altcode directive instructs the compiler to generate alternate code for vectorizedor parallelized loops.
The noaltcode directive disables generation of alternate code.
Scope: This directive affects the compiler only when -Mvect=sse or -Mconcur isenabled on the command line.!pgi$ altcode
Enables alternate code (altcode) generation for vectorized loops. For each loop thecompiler decides whether to generate altcode and what type(s) to generate, whichmay be any or all of: altcode without iteration peeling, altcode with non-temporalstores and other data cache optimizations, and altcode based on array alignmentscalculated dynamically at runtime. The compiler also determines suitable loop countand array alignment conditions for executing the alternate code.
!pgi$ altcode alignmentFor a vectorized loop, if possible, generates an alternate vectorized loop containingadditional aligned moves which is executed if a runtime array alignment test ispassed.
!pgi$ altcode [(n)] concurFor each auto-parallelized loop, generates an alternate serial loop to be executed if theloop count is less than or equal to n. If n is omitted or n is 0, the compiler determinesa suitable value of n for each loop.
!pgi$ altcode [(n)] concurreductionSets the loop count threshold for parallelization of reduction loops to n. For eachauto-parallelized reduction loop, generate an alternate serial loop to be executedif the loop count is less than or equal to n. If n is omitted or n is 0, the compilerdetermines a suitable value of n for each loop.
!pgi$ altcode [(n)] nontemporalFor a vectorized loop, if possible, generates an alternate vectorized loop containingnon-temporal stores and other cache optimizations to be executed if the loop count isgreater than n. If n is omitted or n is 1, the compiler determines a suitable value of nfor each loop. The alternate code is optimized for the case when the data referenced inthe loop does not all fit in level 2 cache.
!pgi$ altcode [(n)] nopeelFor a vectorized loop where iteration peeling is performed by default, if possible,generates an alternate vectorized loop without iteration peeling to be executed if theloop count is less than or equal to n. If n is omitted or n is 1, the compiler determinesa suitable value of n for each loop, and in some cases it may decide not to generate analternate unpeeled loop.
Directives Reference
PVF Reference Guide Version 2017 | 85
!pgi$ altcode [(n)] vectorFor each vectorized loop, generates an alternate scalar loop to be executed if the loopcount is less than or equal to n. If n is omitted or n is 1, the compiler determines asuitable value of n for each loop.
!pgi$ noaltcodeSets the loop count thresholds for parallelization of all innermost loops to 0, anddisables alternate code generation for vectorized loops.
3.1.2. assoc (noassoc)This directive toggles the effects of the -Mvect=noassoc command-line option, anoptimization -M control.
Scope: This directive affects the compiler only when -Mvect=sse is enabled on thecommand line.
By default, when scalar reductions are present the vectorizer may change the order ofoperations, such as dot product, so that it can generate better code. Such transformationsmay change the result of the computation due to roundoff error. The noassoc directivedisables these transformations.
3.1.3. bounds (nobounds)This directive alters the effects of the -Mbounds command line option. This directiveenables the checking of array bounds when subscripted array references are performed.By default, array bounds checking is not performed.
3.1.4. cncall (nocncall)This directive indicates that loops within the specified scope are considered forparallelization, even if they contain calls to user-defined subroutines or functions. Anocncall directive cancels the effect of a previous cncall.
3.1.5. concur (noconcur)This directive alters the effects of the -Mconcur command-line option. The directiveinstructs the auto-parallelizer to enable auto-concurrentization of loops.
Scope: This directive affects the compiler only when -Mconcur is enabled on thecommand line.
If concur is specified, the compiler uses multiple processors to execute loops which theauto-parallelizer determines to be parallelizable. The noconcur directive disables thesetransformations; however, use of concur overrides previous noconcur statements.
Directives Reference
PVF Reference Guide Version 2017 | 86
3.1.6. depchk (nodepchk)This directive alters the effects of the -Mdepchk command line option. When potentialdata dependencies exist, the compiler, by default, assumes that there is a datadependence that in turn may inhibit certain optimizations or vectorizations. nodepchkdirects the compiler to ignore unknown data dependencies.
3.1.7. eqvchk (noeqvchk)The eqvchk directive specifies to check dependencies between EQUIVALENCEassociated elements. When examining data dependencies, noeqvchk directs the compilerto ignore any dependencies between variables appearing in EQUIVALENCE statements.
3.1.8. invarif (noinvarif)This directive has no corresponding command-line option. Normally, the compilerremoves certain invariant if constructs from within a loop and places them outside ofthe loop. The directive noinvarif directs the compiler not to move such constructs. Thedirective invarif toggles a previous noinvarif.
3.1.9. ivdepThe ivdep directive assists the compiler's dependence analysis and is equivalent to thedirective nodepchk.
3.1.10. lstval (nolstval)This directive has no corresponding command-line option. The compiler determineswhether the last values for loop iteration control variables and promoted scalars needto be computed. In certain cases, the compiler must assume that the last values of thesevariables are needed and therefore computes their last values. The directive nolstvaldirects the compiler not to compute the last values for those cases.
3.1.11. optThe opt directive overrides the value specified by the -On command line option.
The syntax of this directive is:!pgi$<scope> opt=<level>
where the optional <scope> is r or g and <level> is an integer constant representingthe optimization level to be used when compiling a subprogram (routine scope) or allsubprograms in a file (global scope).
Directives Reference
PVF Reference Guide Version 2017 | 87
3.1.12. prefetchThe prefetch directive the compiler emits prefetch instructions whereby elementsare fetched into the data cache prior to first use. By varying the prefetch distance,it is sometimes possible to reduce the effects of main memory latency and improveperformance.
The syntax of this directive is:!$mem prefetch <var1>[,<var2>[,...]]
where <varn> is any valid variable, member, or array element reference.
3.1.13. safe_lastvalDuring parallelization, scalars within loops need to be privatized. Problems are possibleif a scalar is accessed outside the loop. If you know that a scalar is assigned on the lastiteration of the loop, making it safe to parallelize the loop, you use the safe_lastvaldirective to let the compiler know the loop is safe to parallelize.
For example, use the following pragma to tell the compiler that for a given loop the lastvalue computed for all scalars make it safe to parallelize the loop:cpgi$l safe_lastval
The command-line option-Msafe_lastval provides the same information for all loopswithin the routines being compiled, essentially providing global scope.
In the following example, the value of t may not be computed on the last iteration of theloop.do i = 1, N if( f(x(i)) > 5.0) then t = x(i) endifenddov = t
If a scalar assigned within a loop is used outside the loop, we normally save the lastvalue of the scalar. Essentially the value of the scalar on the "last iteration" is saved, inthis case when i=N.
If the loop is parallelized and the scalar is not assigned on every iteration, it may bedifficult to determine on what iteration t is last assigned, without resorting to costlycritical sections. Analysis allows the compiler to determine if a scalar is assigned on
Directives Reference
PVF Reference Guide Version 2017 | 88
every iteration, thus the loop is safe to parallelize if the scalar is used later. An exampleloop is:do i = 1, N if( x(i) > 0.0 ) then t = 2.0 else t = 3.0 endif ... y(i) = t ...enddov = t
where t is assigned on every iteration of the loop. However, there are cases where ascalar may be privatizable. If it is used after the loop, it is unsafe to parallelize. Examinethis loop:do i = 1,N if( x(i) > 0.0 ) then t = x(i) ... y(i) = t ... endifenddov = t
where each use of t within the loop is reached by a definition from the same iteration.Here t is privatizable, but the use of t outside the loop may yield incorrect results sincethe compiler may not be able to detect on which iteration of the parallelized loop t isassigned last.
The compiler detects these cases. When a scalar is used after the loop, but is not definedon every iteration of the loop, parallelization does not occur.
3.1.14. tpYou use the directive tp to specify one or more processor targets for which to generatecode.!pgi$ tp [target]...
The tp directive can only be applied at the routine or global level. For moreinformation about these levels, refer to the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
Refer to -tp <target>[,target...] for a list of targets that can be used as parameters to thetp directive.
Directives Reference
PVF Reference Guide Version 2017 | 89
3.1.15. unroll (nounroll)The unroll directive enables loop unrolling while nounroll disables loop unrolling.
The unroll directive has no effect on vectorized loops.
The unroll directive takes arguments c, n and m.
‣ c specifies that c complete unrolling should be turned on or off.‣ n specifies single block loop unrolling.‣ m specifies multi-block loop unrolling.
In addition, a constant may be specified for the c, n and m arguments.
‣ c:v sets the threshold to which c unrolling applies. v is a constant; and a loop whoseconstant loop count is less than or equal to (<=) v is completely unrolled.!pgi$ unroll = c:v
‣ n:v unrolls single block loops v times.!pgi$ unroll = n:v
‣ m:v unrolls single block loops v times.!pgi$ unroll = m:v
The directives unroll and nounroll only apply if-Munroll is selected on the commandline.
3.1.16. vector (novector)The directive novector disables vectorization. The directive vector re-enablesvectorization after a previous novector directive. The directives vector and novector onlyapply if -Mvect has been selected on the command line.
3.1.17. vintr (novintr)The directive novintr directs the vectorizer to disable recognition of vector intrinsics.The directive vintr is re-enables recognition of vector intrinsics after a previous novintrdirective. The directives vintr and novintr only apply if -Mvect has been selected on thecommand line.
3.2. Prefetch Directives and PragmasPrefetch instructions can increase the speed of an application substantially by bringingdata into cache so that it is available when the processor needs it. The PGI prefetchdirective takes the form:
Directives Reference
PVF Reference Guide Version 2017 | 90
The syntax of a prefetch directive in Fortran is as follows:!$mem prefetch <var1>[,<var2>[,...]]
where <varn> is any valid variable, member, or array element reference.
For examples on how to use the prefetch directive, refer to the Prefetch Directivessection of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
3.3. IGNORE_TKR DirectiveThis directive indicates to the compiler to ignore the type, kind, and/or rank (/TKR/)of the specified dummy arguments in an interface of a procedure. The compiler alsoignores the type, kind, and/or rank of the actual arguments when checking all thespecifics in a generic call for ambiguities.
3.3.1. IGNORE_TKR Directive SyntaxThe syntax for the IGNORE_TKR directive is this:!DIR$ IGNORE_TKR [ [(<letter>) <dummy_arg>] ... ]
<letter>is one or any combination of the following:
T – type K – kind R – rank
For example, KR indicates to ignore both kind and rank rules and TKR indicates toignore the type, kind, and rank arguments.
<dummy_arg>if specified, indicates the dummy argument for which TKR rules should be ignored. Ifnot specified, TKR rules are ignored for all dummy arguments in the procedure thatcontains the directive.
3.3.2. IGNORE_TKR Directive Format RequirementsThe following rules apply to this directive:
‣ IGNORE_TKR must not specify dummy arguments that are allocatable, Fortran 90pointers, or assumed-shape arrays.
‣ IGNORE_TKR may appear in the body of an interface block or in the body of amodule procedure, and may specify dummy argument names only.
‣ IGNORE_TKR may appear before or after the declarations of the dummy argumentsit specifies.
‣ If dummy argument names are specified, IGNORE_TKR applies only to thoseparticular dummy arguments.
‣ If no dummy argument names are specified, IGNORE_TKR applies to all dummyarguments except those that are allocatable objects, Fortran 90 pointers, or assumed-shape arrays.
Directives Reference
PVF Reference Guide Version 2017 | 91
3.3.3. Sample Usage of IGNORE_TKR DirectiveConsider this subroutine fragment:subroutine example(A,B,C,D)!DIR$ IGNORE_TKR A, (R) B, (TK) C, (K) D
Table 12 indicates which rules are ignored for which dummy arguments in thepreceding sample subroutine fragment:
Table 12 IGNORE_TKR Example
Dummy Argument Ignored Rules
A Type, Kind and Rank
B Only rank
C Type and Kind
D Only Kind
Notice that no letters were specified for A, so all type, kind, and rank rules are ignored.
3.4. !DEC\$ DirectivesPGI Fortran compilers for Microsoft Windows support directives that help with inter-language calling and importing and exporting routines to and from DLLs. Thesedirectives all take the form:!DEC$ directive
For specific format requirements, refer to the section ‘!DEC$ Directives’ in the PGICompiler User's Guide, https://www.pgroup.com/resources/docs.php.
3.4.1. ALIAS DirectiveThis directive specifies an alternative name with which to resolve a routine.
The syntax for the ALIAS directive is either of the following:!DEC$ ALIAS routine_name , external_name!DEC$ ALIAS routine_name : external_name
In this syntax, external_name is used as the external name for the specifiedroutine_name.
If external_name is an identifier name, the name (in uppercase) is used as the externalname for the specified routine_name. If external_name is a character constant, it isused as-is; the string is not changed to uppercase, nor are blanks removed.
You can also supply an alias for a routine using the ATTRIBUTES directive, described inthe next section:!DEC$ ATTIRIBUTES ALIAS : 'alias_name' :: routine_name
Directives Reference
PVF Reference Guide Version 2017 | 92
This directive specifies an alternative name with which to resolve a routine, as illustratedin the following code fragment that provides external names for three routines. In thisfragment, the external name for sub1 is name1, for sub2 is name2, and for sub3 isname3.subroutine sub!DEC$ alias sub1 , 'name1'!DEC$ alias sub2 : 'name2'!DEC$ attributes alias : 'name3' :: sub3
3.4.2. ATTRIBUTES DirectiveThis directive lets you specify properties for data objects and procedures.
The syntax for the ATTRIBUTES directive is this:!DEC$ ATTRIBUTES <list>
where <list> is one of the following:ALIAS : 'alias_name' :: routine_name
Specifies an alternative name with which to resolve routine_name.C :: routine_name
Specifies that the routine routine_name will have its arguments passed by value.When a routine marked C is called, arguments, except arrays, are sent by value. Forcharacters, only the first character is passed. The standard Fortran calling conventionis pass by reference.
DLLEXPORT :: nameSpecifies that name is being exported from a DLL.
DLLIMPORT :: nameSpecifies that name is being imported from a DLL.
NOMIXED_STR_LEN_ARGSpecifies that hidden lengths are placed in sequential order at the end of the list.
This attribute only applies to routines that are compiled with -Miface=cref orthat use the default Windows calling conventions.
REFERENCE :: nameSpecifies that the argument name is being passed by reference. Often this attribute isused in conjunction with STDCALL, where STDCALL refers to an entire routine; thenindividual arguments are modified with REFERENCE.
STDCALL :: routine_nameSpecifies that routine routine_name will have its arguments passed by value. Whena routine marked STDCALL is called, arguments (except arrays and characters) will besent by value. The standard Fortran calling convention is pass by reference.
VALUE :: nameSpecifies that the argument 'name' is being passed by value.
3.4.3. DECORATE DirectiveThe DECORATE directive specifies that the name specified in the ALIAS directiveshould have the prefix and postfix decorations performed on it that are associated
Directives Reference
PVF Reference Guide Version 2017 | 93
with the calling conventions that are in effect. These declarations are the same onesperformed on the name when ALIAS is not specified.
The syntax for the DECORATE directive is this:!DEC$ DECORATE
When ALIAS is not specified, this directive has no effect.
3.4.4. DISTRIBUTE DirectiveThis directive is front-end based, and tells the compiler at what point within a loop tosplit into two loops.
The syntax for the DISTRIBUTE directive is either of the following:!DEC$ DISTRIBUTE POINT!DEC$ DISTRIBUTEPOINT
Example:subroutine dist(a,b,n) integer i integer n integer a(*) integer b(*) do i = 1,n a(i) = a(i)+2!DEC$ DISTRIBUTE POINT b(i) = b(i)*4 enddoend subroutine
PVF Reference Guide Version 2017 | 94
Chapter 4.RUNTIME ENVIRONMENT
This section describes the programming model supported for compiler code generation,including register conventions and calling conventions for x64 processor-based systemsrunning a Windows operating system.
In this section we sometimes refer to word, halfword, and double word. Theequivalent byte information is word (4 byte), halfword (2 byte), and double word (8byte).
4.1. Win64 Programming ModelThis section defines compiler and assembly language conventions for the use of certainaspects of an x64 processor running a Win64 operating system. These standards mustbe followed to guarantee that compilers, application programs, and operating systemswritten by different people and organizations will work together. The conventionssupported by the Fortran compiler implement the application binary interface (ABI) asdefined in the AMD64 Software Conventions document.
4.1.1. Function Calling SequenceThis section describes the standard function calling sequence, including the stack frame,register usage, and parameter passing.
Register Usage Conventions
Table 13 defines the standard for register allocation. The 64-bit AMD64 and Intel 64architectures provide a number of registers. All the general purpose registers, XMMregisters, and x87 registers are global to all procedures in a running program.
Runtime Environment
PVF Reference Guide Version 2017 | 95
Table 13 Register Allocation
Type Name Purpose
General %rax return value register
%rbx callee-saved
%rcx pass 1st argument to functions
%rdx pass 2nd argument to functions
%rsp stack pointer
%rbp callee-saved; optional stack frame pointer
%rsi callee-saved
%rdi callee-saved
%r8 pass 3rd argument to functions
%r9 pass 4th argument to functions
%r10-%r11 temporary registers; used in syscall/sysret instructions
%r12-r15 callee-saved registers
XMM %xmm0 pass 1st floating point argument; return value register
%xmm1 pass 2nd floating point argument
%xmm2 pass 3rd floating point argument
%xmm3 pass 4th floating point argument
%xmm4-%xmm5 temporary registers
%xmm6-%xmm15 callee-saved registers
In addition to the registers, each function has a frame on the run-time stack. This stackgrows downward from high addresses. Table 14 shows the stack frame organization.
Table 14 Standard Stack Frame
Position Contents Frame
8n-120 (%rbp) argument eightbyte n previous
. . .
-80 (%rbp) argument eightbyte 5
-88 (%rbp) %r9 home
-96 (%rbp) %r8 home
Runtime Environment
PVF Reference Guide Version 2017 | 96
Position Contents Frame
-104 (%rbp) %rdx home
-112 (%rbp) %rcx home
-120 (%rbp) return address current
-128 (%rbp) caller's %rbp
. . .
0 (%rsp) variable size
Key points concerning the stack frame:
‣ The parameter area at the bottom of the stack must contain enough space to holdall the parameters needed by any function call. Space must be set aside for thefour register parameters to be "homed" to the stack even if there are less than fourregister parameters used in a given call.
‣ Sixteen-byte alignment of the stack is required except within a function’s prolog andwithin leaf functions.
All registers on an x64 system are global and thus visible to both a calling and a calledfunction. Registers %rbx, %rsp, %rbp, %rsi, %rdi, %r12, %r13, %r14, and %r15 are non-volatile. Therefore, a called function must preserve these registers’ values for its caller.Remaining registers are scratch. If a calling function wants to preserve such a registervalue across a function call, it must save a value in its local stack frame.
Registers are used in the standard calling sequence. The first four arguments arepassed in registers. Integral and pointer arguments are passed in these general purposeregisters (listed in order): %rcx, %rdx, %r8, %r9. Floating point arguments are passed inthe first four XMM registers: %xmm0, %xmm1, %xmm2, %xmm3. Registers are assignedusing the argument’s ordinal position in the argument list. For example, if a function’sfirst argument is an integral type and its second argument is a floating-point type, thefirst argument will be passed in the first general purpose register (%rcx) and the secondargument will be passed in the second XMM register (%xmm1); the first XMM registerand second general purpose register are ignored. Arguments after the first four arepassed on the stack.
Integral and pointer type return values are returned in %rax. Floating point returnvalues are returned in %xmm0.
Additional registers with assigned roles in the standard calling sequence:%rsp
The stack pointer holds the limit of the current stack frame, which is the addressof the stack’s bottom-most, valid word. The stack pointer should point to a 16-bytealigned area unless in the prolog or a leaf function.
Runtime Environment
PVF Reference Guide Version 2017 | 97
%rbpThe frame pointer, if used, can provide a way to reference the previous frames on thestack. Details are implementation dependent. A function must preserve this registervalue for its caller.
MXCSRThe flags register MXCSR contains the system flags, such as the direction flag and thecarry flag. The six status flags (MXCSR[0:5]) are volatile; the remainder of the registeris nonvolatile.
x87 - Floating Point Control Word (FPCSR)The control word contains the floating-point flags, such as the rounding mode andexception masking. This register is initialized at process initialization time and itsvalue must be preserved.
Signals can interrupt processes. Functions called during signal handling have nounusual restriction on their use of registers. Moreover, if a signal handling functionreturns, the process resumes its original execution path with registers restored to theiroriginal values. Thus, programs and compilers may freely use all registers withoutdanger of signal handlers changing their values.
4.1.2. Function Return Values
Functions Returning Scalars or No Value
‣ A function that returns an integral or pointer value that fits in 64 bits places its resultin %rax.
‣ A function that returns a floating point value that fits in the XMM registers returnsthis value in %xmm0.
‣ A function that returns a value in memory via the stack places the address of thismemory (passed to the function as a "hidden" first argument in %rcx) in %rax.
‣ Functions that return no value (also called procedures or void functions) put noparticular value in any register.
‣ A call instruction pushes the address of the next instruction (the return address)onto the stack. The return instruction pops the address off the stack and effectivelycontinues execution at the next instruction after the call instruction. A functionthat returns a scalar or no value must preserve the caller's registers as previouslydescribed. Further, the called function must remove the return address fromthe stack, leaving the stack pointer (%rsp) with the value it had before the callinstruction was executed.
Functions Returning Structures or Unions
A function can use either registers or the stack to return a structure or union. The sizeand type of the structure or union determine how it is returned. A structure or union
Runtime Environment
PVF Reference Guide Version 2017 | 98
is returned in memory if it is larger than 8 bytes or if its size is 3, 5, 6, or 7 bytes. Astructure or union is returned in %rax if its size is 1, 2, 4, or 8 bytes.
If a structure or union is to be returned in memory, the caller provides space for thereturn value and passes its address to the function as a "hidden" first argument in %rcx.This address will also be returned in %rax.
4.1.3. Argument Passing
Integral and Pointer Arguments
Integral and pointer arguments are passed to a function using the next available registerof the sequence %rcx, %rdx, %r8, %r9. After this list of registers has been exhausted, allremaining integral and pointer arguments are passed to the function via the stack.
Floating-Point Arguments
Float and double arguments are passed to a function using the next available XMMregister of the sequence %xmm0, %xmm1, %xmm2, %xmm3. After this list of registershas been exhausted, all remaining XMM floating-point arguments are passed to thefunction via the stack.
Array, Structure, and Union Arguments
Arrays and strings are passed to functions using a pointer to caller-allocated memory.
Structure and union arguments of size 1, 2, 4, or 8 bytes will be passed as if they wereintegers of the same size. Structures and unions of other sizes will be passed as apointer to a temporary, allocated by the caller, and whose value contains the value of theargument. The caller-allocated temporary memory used for arguments of aggregate typemust be 16-byte aligned.
Passing Arguments on the Stack
Registers are assigned using the argument’s ordinal position in the argument list. Forexample, if a function’s first argument is an integral type and its second argument is afloating-point type, the first argument will be passed in the first general purpose register(%rcx) and the second argument will be passed in the second XMM register (%xmm1);the first XMM register and second general purpose register are ignored. Arguments afterthe first four are passed on the stack; they are pushed on the stack in reverse order, withthe last argument pushed first.
Runtime Environment
PVF Reference Guide Version 2017 | 99
Parameter Passing
Table 15 shows the register allocation and stack frame offsets for the function declarationand call shown in the following example.typedef struct { int i; float f; } struct1; int i; float f; double d; long l; long long ll; struct1 s1; extern void func (int i, float f, struct1 s1, double d, long long ll, long l); func (i, f, s1, d, ll, l);
Table 15 Register Allocation for Example A-4
General Purpose Registers Floating Point Registers Stack Frame Offset
%rcx: i %xmm0: <ignored> 32: ll
%rdx: <ignored> %xmm1: f 40: l
%r8: s1.i, s1.f %xmm2: <ignored>
%r9: <ignored> %xmm3: d
Implementing a Stack
In general, compilers and programmers must maintain a software stack. The stackpointer, register %rsp, is set by the operating system for the application when theprogram is started. The stack must grow downwards from high addresses.
A separate frame pointer enables calls to routines that change the stack pointer toallocate space on the stack at run-time (e.g. alloca). Some languages can also returnvalues from a routine allocated on stack space below the original top-of-stack pointer.Such a routine prevents the calling function from using %rsp-relative addressing to getat values on the stack. If the compiler does not call routines that leave %rsp in an alteredstate when they return, a frame pointer is not needed and is not used if the compileroption -Mnoframe is specified.
The stack must always be 16-byte aligned except within the prolog and within leaffunctions.
Variable Length Parameter Lists
Parameter passing in registers can handle a variable number of parameters. The Clanguage uses a special method to access variable-count parameters. The stdarg.hand varargs.h files define several functions to access these parameters. A C routinewith variable parameters must use the va_start macro to set up a data structure beforethe parameters can be used. The va_arg macro must be used to access the successiveparameters.
Runtime Environment
PVF Reference Guide Version 2017 | 100
For unprototyped functions or functions that use varargs, floating-point argumentspassed in registers must be passed in both an XMM register and its correspondinggeneral purpose register.
C Parameter Conversion
In C, for a called prototyped function, the parameter type in the called function mustmatch the argument type in the calling function.
‣ If the called function is not prototyped, the calling convention uses the types of thearguments but promotes char or short to int, and unsigned char or unsigned short tounsigned int and promotes float to double, unless you use the -Msingle option.
For more information on the -Msingle option, refer to -M Options by Category.‣ If the called function is prototyped, the unused bits of a register containing a char or
short parameter are undefined and the called function must extend the sign of theunused bits when needed.
Calling Assembly Language Programs
C Program Calling an Assembly-language Routine/* File: testmain.c */main() { long l_para1 = 0x3f800000; float f_para2 = 1.0; double d_para3 = 0.5; float f_return; extern float sum_3 (long para1, float para2, double para3); f_return = sum_3(l_para1,f_para2, d_para3); printf("Parameter one, type long = %08x\n",l_para1); printf("Parameter two, type float = %f\n",f_para2); printf("Parameter three, type double = %g\n",d_para3); printf("The sum after conversion = %f\n",f_return);}# File: sum_3.s# Computes ( para1 + para2 ) + para3 .text .align 16 .globl sum_3sum_3: pushq %rbp leaq 128(%rsp), %rbp cvtsi2ss %ecx, %xmm0 addss %xmm1, %xmm0 cvtss2sd %xmm0, %xmm0 addsd %xmm2, %xmm0 cvtsd2ss %xmm0, %xmm0 popq %rbp ret .type sum_3,@function .size sum_3,.-sum_3
4.1.4. Win64 Fortran SupplementSections A3.4.1 through A3.4.4 of the AMD64 Software Conventions for Win64 definethe Fortran supplement. The register usage conventions set forth in that documentremain the same for Fortran.
Runtime Environment
PVF Reference Guide Version 2017 | 101
Fortran Fundamental Types
Table 16 Win64 Fortran Fundamental Types
Fortran Type Size (bytes) Alignment (bytes)
INTEGER 4 4
INTEGER*1 1 1
INTEGER*2 2 2
INTEGER*4 4 4
INTEGER*8 8 8
LOGICAL 4 4
LOGICAL*1 1 1
LOGICAL*2 2 2
LOGICAL*4 4 4
LOGICAL*8 8 8
BYTE 1 1
CHARACTER*n n 1
REAL 4 4
REAL*4 4 4
REAL*8 8 8
DOUBLE PRECISION 8 8
COMPLEX 8 4
COMPLEX*8 8 4
COMPLEX*16 16 8
DOUBLE COMPLEX 16 8
A logical constant is one of:
‣ .TRUE.‣ .FALSE.
The logical constants .TRUE. and .FALSE. are defined to be the four-byte value 1 and 0respectively. A logical expression is defined to be .TRUE. if its least significant bit is 1and .FALSE. otherwise.
Note that the value of a character is not automatically NULL-terminated.
Runtime Environment
PVF Reference Guide Version 2017 | 102
Fortran Naming Conventions
By default, all globally visible Fortran symbol names (subroutines, functions, commonblocks) are converted to lower-case. In addition, an underscore is appended to Fortranglobal names to distinguish the Fortran name space from the C/C++ name space.
Fortran Argument Passing and Return Conventions
Arguments are passed by reference, meaning the address of the argument is passedrather than the argument itself. In contrast, C/C++ arguments are passed by value.
When passing an argument declared as Fortran type CHARACTER, an argumentrepresenting the length of the CHARACTER argument is also passed to the function.This length argument is a four-byte integer passed by value, and is passed at the end ofthe parameter list following the other formal arguments. A length argument is passedfor each CHARACTER argument; the length arguments are passed in the same order astheir respective CHARACTER arguments.
A Fortran function, returning a value of type CHARACTER, adds two arguments to thebeginning of its argument list. The first additional argument is the address of the areacreated by the caller for the return value; the second additional argument is the length ofthe return value. If a Fortran function is declared to return a character value of constantlength, for example CHARACTER*4 FUNCTION CHF(), the second extra parameterrepresenting the length of the return value must still be supplied.
A Fortran complex function returns its value in memory. The caller provides space forthe return value and passes the address of this storage as if it were the first argument tothe function.
Alternate return specifiers of a Fortran function are not passed as arguments by thecaller. The alternate return function passes the appropriate return value back to the callerin %rax.
The handling of the following Fortran 90 features is implementation-defined: internalprocedures, pointer arguments, assumed-shape arguments, functions returning arrays,and functions returning derived types.
Inter-language Calling
Inter-language calling between Fortran and C/C++ is possible if function/subroutineparameters and return values match types. If a C/C++ function returns a value, call itfrom Fortran as a function, otherwise, call it as a subroutine. If a Fortran function hastype CHARACTER or COMPLEX, call it from C/C++ as a void function. If a Fortransubroutine has alternate returns, call it from C/C++ as a function returning int; the valueof such a subroutine is the value of the integer expression specified in the alternateRETURN statement. If a Fortran subroutine does not contain alternate returns, call itfrom C/C++ as a void function.
Runtime Environment
PVF Reference Guide Version 2017 | 103
Table 17 provides the C/C++ data type corresponding to each Fortran data type.
Table 17 Fortran and C/C++ Data Type Compatibility
Fortran Type C/C++ Type Size (bytes)
CHARACTER*n x char x[n] n
REAL x float x 4
REAL*4 x float x 4
REAL*8 x double x 8
DOUBLE PRECISION x double x 8
INTEGER x int x 4
INTEGER*1 x signed char x 1
INTEGER*2 x short x 2
INTEGER*4 x int x 4
INTEGER*8 x long long x 8
LOGICAL x int x 4
LOGICAL*1 x char x 1
LOGICAL*2 x short x 2
LOGICAL*4 x int x 4
LOGICAL*8 x long long x 8
The PGI Compiler User’s Guide contains a table that provides the Fortran and C/C++representation of the COMPLEX type.
Table 18 Fortran and C/C++ Representation of the COMPLEX Type
Fortran Type (lower case) C/C++ Type Size (bytes)
complex x struct {float r,i;} x; 8
float complex x; 8
complex*8 x struct {float r,i;} x; 8
float complex x; 8
double complex x struct {double dr,di;} x; 16
double complex x; 16
complex *16 x struct {double dr,di;} x; 16
Runtime Environment
PVF Reference Guide Version 2017 | 104
Fortran Type (lower case) C/C++ Type Size (bytes)
double complex x; 16
For C/C++, the complex type implies C99 or later.
Arrays
For a number of reasons inter-language function mixing is not recommended for arraysother than single dimensional arrays and square two-dimensional arrays.
‣ C/C++ arrays and Fortran arrays use different default initial array index values. Bydefault, C/C++ arrays start at 0 and Fortran arrays start at 1. However, a Fortranarray can be declared to start at zero.
‣ Fortran and C/C++ arrays use different storage methods. Fortran uses column-majororder and C/C++ use row-major order. For one-dimensional arrays, this poses noproblems. For two-dimensional arrays, where there are an equal number of rowsand columns, row and column indexes can simply be reversed.
Structures, Unions, Maps, and Derived Types.
Fields within Fortran structures and derived types, and multiple map declarationswithin a Fortran union, conform to the same alignment requirements used by Cstructures.
Common Blocks
A named Fortran common block can be represented in C/C++ by a structure whosemembers correspond to the members of the common block. The name of the structure inC/C++ must have the added underscore. Here is an example.
Fortran common block: INTEGER I, J COMPLEX C DOUBLE COMPLEX CD DOUBLE PRECISION D COMMON /COM/ i, j, c, cd, d
C equivalent: extern struct { int i; int j; struct {float real, imag;} c; struct {double real, imag;} cd; double d; } com_;
Runtime Environment
PVF Reference Guide Version 2017 | 105
C++ equivalent: extern "C" struct { int i; int j; struct {float real, imag;} c; struct {double real, imag;} cd; double d; } com_;
The compiler-provided name of the BLANK COMMON block is implementation-specific.
Calling Fortran COMPLEX and CHARACTER functions from C/C++ is not asstraightforward as calling other types of Fortran functions. Additional arguments mustbe passed to the Fortran function by the C/C++ caller. A Fortran COMPLEX functionreturns its value in memory; the first argument passed to the function must containthe address of the storage for this value. A Fortran CHARACTER function adds twoarguments to the beginning of its argument list. The following example of calling aFortran CHARACTER function from C/C++ illustrates these caller-provided extraparameters:CHARACTER*(*) FUNCTION CHF(C1, I)CHARACTER*(*) C1INTEGER I END
extern void chf_();char tmp[10];char c1[9];int i;chf_(tmp, 10, c1, &i, 9);
The extra parameters tmp and 10 are supplied for the return value, while 9 is suppliedas the length of c1.
PVF Reference Guide Version 2017 | 106
Chapter 5.PVF PROPERTIES
There are a number of property pages that are available in a PVF project. These propertypages are grouped into categories that you can access from the Property Page dialog.Further, each of PVF’s property pages contains one or more properties, or configurationoptions. The set of categories and property pages available vary, depending on the typeof project.
The properties in a PVF project are divided into the following categories:
‣ General‣ Debugging‣ Fortran‣ Linker
‣ Librarian‣ Resources‣ Build Events‣ Custom Build Step
This section contains descriptions of each of PVF’s property pages, and detaileddescriptions of the properties, organized as you would see them in the Property Pagedialog: by category and property page.
Tip The Fortran, Linker, and Librarian categories contain a Command Line propertypage where you can see the command line derived from the properties in thatcategory. Options that are not supported by the PVF property pages can be added tothe command line from this property page by entering them in the Additional Optionsfield.
5.1. General Property PageThis section contains the properties that are included on the General property page.
5.1.1. General
PVF Properties
PVF Reference Guide Version 2017 | 107
5.1.2. Output DirectoryUse this property to specify a relative path to the output file directory. This directory iswhere the project’s output files are built.
5.1.3. Intermediate DirectoryUse this property to specify a relative path to the intermediate file directory. Thisdirectory is where the intermediate files (i.e., object files) are created when the project isbuilt.
5.1.4. Extensions to Delete on CleanUse this property to specify which files in the intermediate directory should be deletedwhen the project is cleaned or before it is rebuilt. This property uses a semi-colon-delimited wildcard specification for the files.
5.1.5. Configuration TypeUse this property to change the output type that the project produces.
When you create a project, you specify the type of output that the project produces:executable, static library, or dynamic library. If you want to change the output type, usethis property to do so.
5.1.6. Build Log FileUse this property to specify the build log file that is produced when the project is built.
5.1.7. Build Log LevelUse this property to specify the level of detail to be included in the build log file.
Any setting above Default can produce large amounts of output and may potentiallyslow down the building of your project.
5.2. Debugging Property PageThis section contains the properties that are included on the Debugging property page.
5.2.1. Debugging
5.2.2. Application CommandUse this property to specify the application to execute when you select Start Debuggingor Start Without Debugging from the Debug menu.
PVF Properties
PVF Reference Guide Version 2017 | 108
‣ If the Startup Project in your solution is a PVF project that builds an executable,there is probably no need to change this property.
‣ If the Startup Project in your solution is a PVF project that builds a DLL or staticlibrary, you must use the Command property to specify an application to executewhen you run (with or without debugging).
To use the PVF debug engine, the Startup Project must be a PVF project. If, forexample, your main executable is built by a Visual C++ project that links againsta PVF project, you would designate the PVF project as the Startup Project; and inits Debugging | Application Command property, you would specify the path to theexecutable built by the Visual C++ project.
Tip The Startup Project is the project listed in boldface in the solution explorer.You can change the Startup Project by right-clicking on any project in the solutionexplorer and selecting Set as Startup Project from the context menu.
5.2.3. Application ArgumentsUse this property to pass command line arguments to the application when it is run ordebugged.
5.2.4. EnvironmentUse this property to specify any environment variables to set for the application whenit runs. One common use of this property is to augment the PATH environment variable.For example, if the application requires DLLs to run but the general environment isnot set to find these, the path to these DLLs could be added to the PATH environmentvariable.
For more information on PATH, refer to the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
If the Merge Environment property is set to Yes, then the contents of the Environmentproperty are merged with the existing environment when the application is run ordebugged.
5.2.5. Merge EnvironmentUse this property to merge the environment variables in the Environment property withthe existing environment when the application is run or debugged. To do this, set theMerge Environment property to Yes.
5.2.6. Accelerator ProfilingUse this property to generate accelerator profiling information at runtime. To do this, setthe Accelerator Profiling property to Yes.
Setting this property to Yes sets the PGI_ACC_TIME environment variable to 1.
PVF Properties
PVF Reference Guide Version 2017 | 109
5.2.7. MPI DebuggingUse this property to enable MPI debugging and select local MPI debugging.
The value selected for this property determines which properties are displayedfollowing it on the Debugging property page.
Important If you change the value of this property and the displayed properties donot change, be sure to click Apply in the property page dialog box.
‣ When MPI Debugging is set to Disabled, the application is run or debugged in serialmode.
‣ When MPI Debugging is set to Local, the application is run or debugged usingmpiexec. All processes launched are local to the system on which the application isrun.
5.2.8. Working Directory
[Serial]
Use this property to specify the application's working directory when it is run ordebugged serially. By default, the working directory is set to the solution directory.
This property is displayed when the MPI Debugging property is set to Disabled.
5.2.9. Number of Processes
[Local MPI]
Use this property to specify the number of MPI processes to use when the application isrun or debugged. The number of processes is passed to mpiexec using the -n option.
This property is displayed when the MPI Debugging property is set to Local.
5.2.10. Working Directory
[Local MPI]
Use this property to specify the application's working directory when it is run ordebugged using local MPI. By default, the working directory is set to the solutiondirectory.
This property is displayed when the MPI Debugging property is set to Local.
PVF Properties
PVF Reference Guide Version 2017 | 110
5.2.11. Additional Arguments: mpiexec
[Local MPI]
Use this property to specify additional arguments to be passed to mpiexec when theapplication is run or debugged.
This property is displayed when the MPI Debugging property is set to Local.
5.2.12. Location of mpiexec
[Local MPI]
Use this property to override the default path to mpiexec as specified in the systemPATH variable.
This property is displayed when the MPI Debugging property is set to Local.
5.3. Fortran Property PagesThis section contains the property pages that are included in the Fortran category. Thiscategory is further divided into the following property pages, displayed in the followingorder:
‣ General‣ Optimization‣ Preprocessing‣ Code Generation
‣ Language‣ Floating Point Options‣ External Procedures‣ Target Processors
‣ Target Accelerators‣ Diagnostics‣ Profiling‣ Command Line
The following sections describe the properties available on each property page.
5.4. Fortran | GeneralThe following properties are available from the Fortran | General property page.
5.4.1. Display Startup BannerUse this property to determine whether to display the compiler’s startup banner duringcompilation.
Changing the property to Yes adds the -V switch to the compilation line, which causesthe compiler to display the startup banner during compilation.
For more information on -V, refer to -V[release_number].
PVF Properties
PVF Reference Guide Version 2017 | 111
5.4.2. Additional Include DirectoriesUse this property to add one or more directories to the compiler’s include path.
For every path that is added to this property, PVF adds -I<path> to the compilationline.
There are two ways to add directories to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each directory.‣ Click the ellipsis (‘...’) button in the property page box to open the Additional Include
Directories dialog box.
Enter each directory on its own line in this box. Do not use semi-colons to separatedirectories; the semi-colons are added automatically when the box is closed.
This property is also available from the Fortran | Preprocessing Property page.
5.4.3. Module PathUse this property to specify the location of module (.mod) files.
For every directory that is added to this property, PVF adds -module <dir> to thecompilation line, causing the compiler to search each listed directory for modules duringcompilation.
The first directory in the list is also the module output directory, which is where PVFputs all module files created when the project is built.
There are two ways to add directories to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each path.‣ Click the ellipsis (‘...’) button in the property page box to open the Module Path
dialog box.
Enter each directory on its own line in this box. Do not use semi-colons to separateentries; the semi-colons are added automatically when the box is closed.
5.4.4. Object File NameUse of this property depends on whether it is being applied to a file or a project:
‣ File level: Use this property to set the name of the object file. Setting the name addsthe -o switch to the compilation line.
For more information on -o, refer to -o.
PVF Properties
PVF Reference Guide Version 2017 | 112
‣ Project level: Use this property to set the location of the object files created by abuild.
To change the default location for the object files, specify a different directory namefor this property.
You must append a backslash (\) to the directory path or the value of thisproperty will be interpreted as a file.
5.4.5. Debug Information FormatUse this property to specify whether the compiler should generate debug informationand if so, in what format.
‣ The richest debugging experience is obtained when this option is set to "Full DebugInformation (-g).’
‣ If you are debugging a project built with optimizations, you may want to select"Full Debug Information with Full Optimization (-gopt)." This selection prevents thegeneration of debug information from affecting optimizations.
For more information on -g, refer to -g. For more information on -gopt, refer to -gopt.
5.4.6. OptimizationUse this property to select the overall code optimization.
This property can be set to one of the following values:
‣ No Optimization - the default value for Debug configurations.‣ Maximize Speed - the default value for Release Configurations.‣ Maximize Speed Across the Whole Program
This property is also available from the Fortran | Optimization Property page.
5.5. Fortran | OptimizationThe following properties are available from the Fortran | Optimization property page.
5.5.1. OptimizationUse this property to select the overall code optimization.
This property can be set to one of the following values:
‣ No Optimization - the default value for Debug configurations.‣ Maximize Speed - the default value for Release Configurations.
PVF Properties
PVF Reference Guide Version 2017 | 113
‣ Maximize Speed Across the Whole Program
This property is also available from the Fortran | General Property page.
5.5.2. Global OptimizationsUse this property to set the compiler’s global optimization level.
Setting this property adds one of the -O options to the compilation line.
For more information on -O, refer to -O<level>.
5.5.3. VectorizationUse this property to specify the type of vectorization to perform.
The PVF compilers use the -Mvect options to vectorize code that is vectorizable. Selectthe appropriate vectorization from these options:
‣ Default: Accepts the default vectorization.‣ Enable Vectorization: Enables vectorization by adding the -Mvect switch to the
PVF compilation and link lines.‣ Vectorize using SSE instructions: Enables vectorization using SSE instructions by
adding the -Mvect=sse switch to the PVF compilation line.‣ Vectorize using SIMD instructions: Enables vectorization using SIMD instructions
and data, by adding the -Mvect=simd switch to the PVF compilation line.‣ Vectorize using 128-bit SIMD instructions: Enables vectorization using SIMD 128-
bit instructions and data, by adding the -Mvect=simd:128 switch to the PVFcompilation line.
‣ Vectorize using 256-bit SIMD instructions: Enables vectorization using SIMD 256-bit instructions and data, by adding the -Mvect=simd:256 switch to the PVFcompilation line.
For more information on -Mvect, refer to Optimization Controls.
5.5.4. InliningUse this property to enable inlining of certain subprograms.
Setting this property to Yes adds the -Minline switch to the compilation commandline.
For more information on -Minline, refer to -Minline[=option[,option,...]].
5.5.5. Use Frame PointerUse this property to specify whether to generate code that uses a frame pointer.
Setting this property to Yes adds the -Mframe switch to the compilation command lineand PVF compilers generate code that uses a frame pointer.
PVF Properties
PVF Reference Guide Version 2017 | 114
Setting this property to No, the default, adds the -Mnoframe switch to the compilationcommand line and PVF compilers generate code that does not use frame pointers.
For more information on -Mframe, refer to Optimization Controls.
5.5.6. Loop Unroll CountUse this property to select the appropriate value for unrolling.
Loop unrolling is a common optimization. This property allows you to specify unrollingby two or four. Using this option adds the -Munroll option to the compilation line.
For more information on -Munroll, refer to Optimization Controls.
5.5.7. Auto-ParallelizationUse this property to auto-parallelize code that is parallelizable. Using this option addsthe -Mconcur option to the compilation line.
For more information on -Mconcur, refer to Optimization Controls.
5.6. Fortran | PreprocessingThe following properties are available from the Fortran | Preprocessing Property page.
5.6.1. Preprocess Source FileUse this property to specify whether the compiler should preprocess source files.
Setting this property to Yes adds the -Mpreprocess switch to the compilationcommand line.
For more information on -Mpreprocess, refer to Miscellaneous Controls.
5.6.2. Additional Include DirectoriesUse this property to add one or more directories to the compiler’s include path.
For every path that is added to this property, PVF adds -I<path> to the compilationline.
There are two ways to add directories to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each directory.‣ Click the ellipsis (‘...’) button in the property page box to open the Additional Include
Directories dialog box.
Enter each directory on its own line in this box. Do not use semi-colons to separatedirectories; the semi-colons are added automatically when the box is closed.
PVF Properties
PVF Reference Guide Version 2017 | 115
For more information on -I<path>, refer to -I.
This property is also available from the Fortran | General Property page.
5.6.3. Ignore Standard Include PathUse this property to specify whether the preprocessor should ignore the standardinclude path.
Setting this property to Yes adds the -Mnostdinc switch to the compilation commandline.
For more information on -Mnostdinc, refer to Environment Controls.
5.6.4. Preprocessor DefinitionsUse this property to add one or more preprocessor definitions to compilation.
For every definition that is added to this property, PVF adds -D<definition> to thecompilation line.
There are two ways to add definitions to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each definition.
For example, DEF1;DEF2=2 defines DEF1, and defines DEF2 and initializes it to 2.‣ Click the ellipsis (‘...’) button in the property page box to open the Preprocessor
Definitions dialog box.
Enter each definition on its own line in this box. Do not use semi-colons to separatedefinitions; the semi-colons are added automatically when the box is closed.
For more information on -D<definition>, refer to -D.
5.6.5. Undefine Preprocessor DefinitionsUse this property to undefine one or more preprocessor definitions.
For every definition that is added to this property, PVF adds -U<definition> to thecompilation line.
There are two ways to add definitions to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each definition.
For example, DEF1;DEF2 undefines DEF1 and DEF2.‣ Click the ellipsis (‘...’) button in the property page box to open the Undefine
Preprocessor Definitions dialog box.
Enter each definition on its own line in this box. Do not use semi-colons to separatedefinitions; the semi-colons are added automatically when the box is closed.
PVF Properties
PVF Reference Guide Version 2017 | 116
For more information on -U<definition>, refer to -U.
5.7. Fortran | Code GenerationThe following properties are available from the Fortran | Code Generation propertypage.
5.7.1. Runtime LibraryUse this property to specify the type of runtime libraries to use during linking.
Default: Depends on the project:
‣ For executable and static library projects: multi-threaded static libraries.
Using this option adds the -Bstatic option to the compilation line. This choicecorresponds to Microsoft’s /MT compilation option.
For more information on -Bstatic, refer to -Bstatic.‣ For dynamic-link library projects: multi-threaded DLL libraries.
Using this option adds the -Bdynamic option to the compilation line. This choicecorresponds to Microsoft’s /MD compilation option.
For more information on -Bdynamic, refer to -Bdynamic.
It is important to keep the type of runtime libraries consistent within a solution. PVFprojects that build DLLs should link to the multi-threaded DLL runtime, and projectsthat link to these PVF DLLs should also use the multi-threaded DLL runtime.
5.8. Fortran | LanguageThe following properties are available from the Fortran | Language property page.
5.8.1. Fortran DialectUse this property to select the Fortran dialect to use during compilation.
PVF supports two Fortran language dialects: Fortran 95 and FORTRAN 77. The dialectdetermines which PGI compiler driver is used during compilation.
‣ Default: The dialect is set to Fortran 95, even for fixed-format .f files, and thepgfortran driver is used.
‣ Fortran 77: Use the pgf77 driver. You can select the FORTRAN 77 dialect at theproject or file level.
5.8.2. Treat Backslash as CharacterUse this property to specify how the compilers should treat the backslash (\) character.
PVF Properties
PVF Reference Guide Version 2017 | 117
Default: PVF treats the backslash (\) as a regular character.
This default action is equivalent to adding the -Mbackslash switch to compilation.
If you want the backslash character to be treated as an escape character, which is how Cand C++ compilers handle backslashes, set this property to No.
For more information on -Mbackslash, refer to Fortran Language Controls.
5.8.3. Extend Line LengthUse this property to extend the line length for fixed-format Fortran files to 132characters.
Fixed-format Fortran files limit the accepted line length to 72 characters. To extend theline length for these types of files to 132 characters, set this property to Yes, which addsthe -Mextend switch to the PVF compilation line.
For more information on -Mextend, refer to Fortran Language Controls.
5.8.4. Enable OpenMP DirectivesUse this property to enable OpenMP 3.0 directives.
Setting this property to Yes adds the -mp switch to the PVF compilation and link lines.
For more information on -mp, refer to -mp.
5.8.5. Enable OpenACC DirectivesUse this property to enable OpenACC directives.
Setting this property to Yes adds the -acc switch to the PVF compilation and link linesand activates access to these additional properties:
OpenACC AutoparallelizationOpenACC RequiredOpenACC RoutineseqOpenACC WaitOpenACC Conformance LevelOpenACC Sync
For more information on -acc, refer to -acc.
5.8.6. OpenACC AutoparallelizationWhen Enable OpenACC Directives is set to Yes, use this property to control loopautoparallelization within acc parallel.
‣ Default: Allows the compiler to control loop autoparallelization within acc parallel.This selection adds no additional sub-options to -acc.
‣ Yes: Directs the compiler to enable loop autoparallelization within an OpenACCparallel region (-acc=autopar).
PVF Properties
PVF Reference Guide Version 2017 | 118
‣ No: Directs the compiler to disable loop autoparallelization within an OpenAccparallel region (-acc=noautopar).
5.8.7. OpenACC RequiredWhen Enable OpenACC Directives is set to Yes, use this property to control thecompiler’s behavior when it is unable to accelerate a compute region.
‣ Default: Use the compiler defaults for handling instances where compute regionscannot be accelerated. This selection adds no additional sub-options to -acc.
‣ Yes: Directs the compiler to stop compilation with an error when it cannot acceleratea compute region (-acc=required).
‣ No: Directs the compiler to issue warnings when it cannot accelerate a computeregion; compilation does not stop but accelerator kernels are not generated (-acc=norequired).
5.8.8. OpenACC RoutineseqWhen Enable OpenACC Directives is set to Yes, use this property to compile everyroutine for the device.
‣ Default: Uses compiler defaults handling compile every routine for the device. Thisselection adds no additional sub-options to -acc.
‣ Yes: Enables compiling every routine for the device by adding -acc=routineseqswitch to the PVF compilation and link lines.
‣ No: Disables compiling every routine for the device by adding -acc=noroutineseq switch to the PVF compilation and link lines.
5.8.9. OpenACC WaitWhen Enable OpenACC Directives is set to Yes, use this property to wait for eachdevice kernel to finish.
‣ Default: Uses compiler defaults handling wait for each device kernel to finish. Thisselection adds no additional sub-options to -acc.
‣ Yes: Enables wait for each device kernel to finish by adding -acc=wait switch tothe PVF compilation and link lines.
‣ No: Disables wait for each device kernel to finish by adding -acc=nowait switchto the PVF compilation and link lines.
5.8.10. OpenACC Conformance LevelWhen Enable OpenACC Directives is set to Yes, use this property to leverage thecompiler’s detection of extensions to standard OpenACC directives.
‣ Default: When non-OpenACC accelerator directives are found, they are ignored..‣ Strict: Add -acc=strict to the PVF compilation and link lines. The compiler emits
a warning when a non-OpenACC accelerator directive is found.‣ Very Strict: Add -acc=strict to the PVF compilation and link lines. The compiler
stops with an error when a non-OpenACC accelerator directive is found.
PVF Properties
PVF Reference Guide Version 2017 | 119
5.8.11. OpenACC SyncWhen Enable OpenACC Directives is set to Yes, use this property to ignore asyncclauses.
Setting this property to Yes adds the -acc=sync switch to the PVF compilation andlink lines.
5.8.12. MPIUse this property to enable compilation and linking using the Microsoft MPI headersand libraries.
Setting this property to Microsoft MPI adds the -Mmpi=msmpi switch to the PVFcompilation and link lines.
5.8.13. Enable CUDA FortranUse this property to enable CUDA Fortran.
Setting this property to Yes adds the -Mcuda switch to the PVF compilation and linklines and activates access to these additional properties:
CUDA Fortran Register LimitCUDA Fortran Use Fused Multiply-AddsCUDA Fortran Use Fast Math LibraryCUDA Fortran DebugCUDA Fortran Line InformationCUDA Fortran Use LLVM Back EndCUDA Fortran UnrollCUDA Fortran Flush to ZeroCUDA Fortran ToolkitCUDA Fortran Compute CapabilityCUDA Fortran Keep BinaryCUDA Fortran Keep Kernel SourceCUDA Fortran Keep PTXCUDA Fortran Keep PTXASCUDA Fortran Generate RDCCUDA Fortran EmulationCUDA Fortran Madconst
Important If you select Yes and the additional properties do not appear, click Applyin the Property page dialog.
For more information on -Mcuda, refer to Optimization Controls.
PVF Properties
PVF Reference Guide Version 2017 | 120
5.8.14. CUDA Fortran Register LimitWhen Enable CUDA Fortran is set to Yes, use this property to specify the number ofregisters to use on the GPU.
Setting this property to an integer value, n, adds the -Mcuda=maxregcount:n switchto the PVF compilation and link lines.
Leaving this property blank indicates no limit to the number of registers to use on theGPU.
For more information on -Mcuda, refer to Optimization Controls.
5.8.15. CUDA Fortran Use Fused Multiply-AddsWhen Enable CUDA Fortran is set to Yes, use this property to control the generation offused multiply-add (FMA) instructions.
‣ Default: Allows the compiler to control generation of FMA instructions. Thisselection adds no additional sub-options to -Mcuda.
‣ Yes: Enables generation of FMA instructions by adding -Mcuda=fma switch to thePVF compilation and link lines.
‣ No: Disables generation of FMA instructions by adding -Mcuda=nofma switch tothe PVF compilation and link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.16. CUDA Fortran Use Fast Math LibraryWhen Enable CUDA Fortran is set to Yes, use this property to use routines from the fastmath library.
Setting this property to Yes adds the -Mcuda=fastmath switch to compilation andlinking.
For more information on -Mcuda, refer to Optimization Controls.
5.8.17. CUDA Fortran DebugWhen Enable CUDA Fortran is set to Yes, use this property to control generatation ofGPU debug information.
‣ Default: Allows the compiler to control generatation of GPU debug information.This selection adds no additional sub-options to -Mcuda.
‣ Yes: Enables generatation of GPU debug information by adding the -Mcuda=debugswitch to the PVF compilation and link lines.
‣ No: Disables generatation of GPU debug information by adding the-Mcuda=nodebug switch to the PVF compilation and link lines.
For more information on -Mcuda, refer to Optimization Controls.
PVF Properties
PVF Reference Guide Version 2017 | 121
5.8.18. CUDA Fortran Line InformationWhen Enable CUDA Fortran is set to Yes, use this property to control generatation ofGPU line information.
‣ Default: Allows the compiler to control generatation of GPU line information. Thisselection adds no additional sub-options to -Mcuda.
‣ Yes: Enables generatation of GPU line information by adding the-Mcuda=lineinfo switch to the PVF compilation and link lines.
‣ No: Disables generatation of GPU line information by adding the-Mcuda=nolineinfo switch to the PVF compilation and link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.19. CUDA Fortran Use LLVM Back EndWhen Enable CUDA Fortran is set to Yes, use this property to control using LLVM backend.
‣ Default: Allows the compiler to control using LLVM back end. This selection addsno additional sub-options to -Mcuda.
‣ Yes: Use LLVM back end by adding the -Mcuda=llvm switch to the PVFcompilation and link lines.
‣ No: Use CUDA C back end by adding the -Mcuda=nollvm switch to the PVFcompilation and link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.20. CUDA Fortran UnrollWhen Enable CUDA Fortran is set to Yes, use this property to control automatic innerloop unrolling.
‣ Default: Allows the compiler to control automatic inner loop unrolling. Thisselection adds no additional sub-options to -Mcuda.
‣ Yes: Enables automatic inner loop unrolling by adding the -Mcuda=unroll switchto the PVF compilation and link lines.
‣ No: Disables automatic inner loop unrolling by adding the -Mcuda=nounrollswitch to the PVF compilation and link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.21. CUDA Fortran Flush to ZeroWhen Enable CUDA Fortran is set to Yes, use this property to control flush-to-zeromode for floating point computations on in GPU code generated for CUDA Fortrankernels.
‣ Default: Accepts the default handling of floating point computations in the GPUcode generated for CUDA Fortran kernels.
PVF Properties
PVF Reference Guide Version 2017 | 122
‣ Yes: Enables flush-to-zero mode by adding the -Mcuda=flushz switch to the PVFcompilation and link lines.
‣ No: Disables flush-to-zero mode by adding the -Mcuda=noflushz switch to thePVF compilation and link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.22. CUDA Fortran ToolkitWhen Enable CUDA Fortran is set to Yes, use this property to specify the version of theCUDA toolkit that is targeted by the compilers.
‣ Default: The compiler selects the default CUDA toolkit version.‣ 7.5: Use the default version 7.5 of the CUDA toolkit. This selection adds the
-Mcuda=cuda7.5 switch to the PVF compilation and link lines.‣ 8.0: Use version 8.0 of the CUDA toolkit. This selection adds the -Mcuda=cuda8.0
switch to the PVF compilation and link lines.
pgaccelinfo prints the driver version as the first line of output.
For a 7.5 driver: CUDA Driver Version 7050For a 8.0 driver: CUDA Driver Version 8000
For more information on -Mcuda, refer to Fortran Language Controls.
5.8.23. CUDA Fortran Compute CapabilityWhen Enable CUDA Fortran is set to Yes, use this property to either automaticallygenerate code compatible with all applicable compute capabilities, or to direct thecompiler to use a manually-selected set.
Select either Automatic or Manual.
‣ Automatic: Let the compiler generate code for all applicable compute capabilities.This is the default.
‣ Manual: Choose one or more compute capabilities to target. The compiler generatescode for each capability specified.
If you select Manual, then you can select any or all of the following computecapabilities that are described in the next sections.
CUDA Fortran FermiCUDA Fortran Fermi+CUDA Fortran KeplerCUDA Fortran Kepler+
Important If you select Manual and the additional properties do not appear, clickApply in the Property page dialog.
For more information on -Mcuda, refer to Optimization Controls.
PVF Properties
PVF Reference Guide Version 2017 | 123
5.8.24. CUDA Fortran FermiWhen Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is setto Manual, use this property to generate code for the Fermi architecture.
Setting this property to Yes adds the -Mcuda=fermi switch to the PVF compilationand link lines.
5.8.25. CUDA Fortran Fermi+When Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is setto Manual, use this property to generate code for Fermi architecture and above.
Setting this property to Yes adds the -Mcuda=fermi+ switch to the PVF compilationand link lines.
5.8.26. CUDA Fortran KeplerWhen Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is setto Manual, use this property to generate code for the Kepler architecture.
Setting this property to Yes adds the -Mcuda=kepler switch to the PVF compilationand link lines.
5.8.27. CUDA Fortran Kepler+When Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is setto Manual, use this property to generate code for Kepler architecture and above.
Setting this property to Yes adds the -Mcuda=kepler+ switch to the PVF compilationand link lines.
5.8.28. CUDA Fortran Keep BinaryUse this property to keep the CUDA binary (.bin) file.
Setting this property to Yes adds the -Mcuda=keepbin switch to the PVF compilationand link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.29. CUDA Fortran Keep Kernel SourceWhen Enable CUDA Fortran is set to Yes, use this property to keep the kernel sourcefiles.
Setting this property to Yes adds the -Mcuda=keepgpu switch to the PVF compilationand link lines.
For more information on -Mcuda, refer to Optimization Controls.
PVF Properties
PVF Reference Guide Version 2017 | 124
5.8.30. CUDA Fortran Keep PTXWhen Enable CUDA Fortran is set to Yes, use this property to keep the portableassembly (.ptx) file for the GPU code.
Setting this property to Yes adds the -Mcuda=keepptx switch to the PVF compilationand link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.31. CUDA Fortran Keep PTXASUse this property to show PTXAS informational messages during compilation.
Setting this property to Yes adds the -Mcuda=ptxinfo switch to the PVF compilationand link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.32. CUDA Fortran Generate RDCUse this property to generate relocatable device code (-Mcuda=rdc).
Setting this property to Yes adds the -Mcuda=rdc switch to the PVF compilation andlink lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.33. CUDA Fortran EmulationWhen Enable CUDA Fortran is set to Yes, use this property to enable CUDA Fortranemulation mode.
Setting this property to Yes adds the -Mcuda=emu switch to the PVF compilation andlink lines.
For more information on -Mcuda, refer to Optimization Controls.
5.8.34. CUDA Fortran MadconstWhen Enable CUDA Fortran is set to Yes, use this property to control putting modulearray descriptors in CUDA constant memory.
Setting this property to Yes adds the -Mcuda=madconst switch to the PVF compilationand link lines.
For more information on -Mcuda, refer to Optimization Controls.
5.9. Fortran | Floating Point OptionsThe following properties are available from the Fortran | Floating Point Optionsproperty page.
PVF Properties
PVF Reference Guide Version 2017 | 125
5.9.1. Floating Point Exception HandlingUse this property to enable floating point exceptions.
Setting this property to Yes adds the -Ktrap=fp option to compilation.
For more information on -Ktrap, refer to -K<flag>.
5.9.2. Floating Point ConsistencyUse this property to enable relaxed floating point accuracy in favor of speed.
Setting this property to Yes adds the -Mfprelaxed option to compilation.
For more information on -Mfprelaxed, refer to Optimization Controls.
5.9.3. Flush Denormalized Results to ZeroUse this property to specify how to handle denormalized floating point results.
‣ Default: Accepts the default handling of denormalized floating point results.‣ Yes: Enables SSE flush-to-zero mode using the -Mflushz compilation option.‣ No: Disables SSE flush-to-zero mode using the -Mnoflushz compilation option.
For more information on -M[no]flushz, refer to Code Generation Controls.
5.9.4. Treat Denormalized Values as ZeroUse this property to specify how to treat denormalized floating point values.
‣ Default: Accept the default treatment of denormalized floating point values.‣ Yes: Enable the treatment of denormalized floating point values as zero using the
-Mdaz compilation option.‣ No: Disable the treatment of denormalized floating point values as zero using the
-Mnodaz compilation option.
For more information on -M[no]daz, refer to Code Generation Controls.
5.9.5. IEEE ArithmeticUse this option to specify IEEE floating point arithmetic.
‣ Default: Accept the default floating point arithmetic.‣ Yes: Enable IEEE floating point arithmetic using the -Kieee compilation option.‣ No: Disable IEEE floating point arithmetic using the -Knoieee compilation option.
For more information on -K[no]ieee, refer to -K<flag>.
PVF Properties
PVF Reference Guide Version 2017 | 126
5.10. Fortran | External ProceduresThe following properties are available from the Fortran | External Procedures propertypage.
5.10.1. Calling ConventionUse this property to specify an alternate Fortran calling convention.
‣ Default: Accept the default calling convention.‣ C By Reference: Use the CREF calling convention. Adds -Miface=cref to
compilation. On x64 platforms, no trailing underscores are used with this option andthis option also causes Fortran externals to be uppercase and lengths of characterarguments to be put at the end of the argument list.
For more information on -Miface, refer to Miscellaneous Controls.
5.10.2. String Length ArgumentsUse this property to change where string length arguments are placed in the argumentlist.
‣ Default: Use the calling convention's default placement for passing string lengtharguments.
‣ After Every String Argument: Lengths of character arguments areplaced immediately after their corresponding argument. This option adds-Miface=mixed_str_len_arg to compilation.
‣ After All Arguments: Places lengths of character arguments at the end of theargument list. This option adds -Miface=nomixed_str_len_arg to thecompilation.
The After Every String Argument and After All Arguments options only have an effectwhen using the C By Reference calling convention.
For more information on -Miface, refer to Miscellaneous Controls.
5.10.3. Case of External NamesUse this property to specify the case used for Fortran external names.
‣ Default: Use the calling convention's default case for external names.‣ Lower Case: Make Fortran external names lower case. This option adds
-Mnames=lowercase to the compilation.
PVF Properties
PVF Reference Guide Version 2017 | 127
‣ Upper Case: Make Fortran external names upper case. This option adds-Mnames=uppercase to the compilation.
The Lower Case and Upper Case options only have an effect when using the C ByReference calling convention.
5.11. Fortran | LibrariesThe Fortran | Libraries property page contains properties that make it easier to usethird-party libraries. To use these libraries, however, the appropriate binaries, such as.lib and .dll files, must be installed on your system.
5.11.1. Use MKLUse this property to build for and link against the Intel Math Kernel Library (MKL),which is available from Intel.
‣ Yes: Use the Intel Math Kernel Library when building and linking programs.‣ No: Do not use the Intel Math Kernel Library when building and linking programs.
5.12. Fortran | Target ProcessorsThe properties that are available from the Fortran | Target Processors property pagedepend on the platform you are using. The platform selection box in the center of theProperty Pages dialog indicates the platform: x64.
x64 PlatformYou can target multiple processors for optimization on the x64 platform.
The Target Processors properties add the -tp=<target> option to compilation. Formore information on the -tp switch referenced throughout the following descriptions,refer to -tp [k8-64].
5.12.1. AMD AthlonUse this property to optimize for AMD Athlon64, AMD Opteron and compatibleprocessors.
x64: Setting this property to Yes adds the -tp=k8-64 switch to compilation.
5.12.2. AMD BarcelonaUse this property to optimize for AMD Opteron/Quadcore and compatible processors.
x64: Setting this property to Yes adds the -tp=barcelona-64 switch to compilation.
PVF Properties
PVF Reference Guide Version 2017 | 128
5.12.3. AMD BulldozerUse this property to optimize for AMD Bulldozer and compatible processors.
x64: Setting this property to Yes adds the -tp=bulldozer-64 switch to compilation.
5.12.4. AMD IstanbulUse this property to optimize for AMD Istanbul processor-based systems.
x64: Setting this property to Yes adds the -tp=istanbul-64 switch to compilation.
5.12.5. AMD PiledriverUse this property to optimize for AMD Piledriver processor-based systems.
x64: Setting this property to Yes adds the -tp=piledriver-64 switch to compilation.
5.12.6. AMD ShanghaiUse this property to optimize for AMD Shanghai processor-based systems.
x64: Setting this property to Yes adds the -tp=shanghai-64 switch to compilation.
5.12.7. Intel Core 2Use this property to optimize for Intel Core 2 Duo and compatible processors.
x64: Setting this property to Yes adds the -tp=core2-64 switch to compilation.
5.12.8. Intel Core i7Use this property to optimize for Intel Core i7 (Nehalem) processor-based systems.
x64: Setting this property to Yes adds the -tp=nehalem-64 switch to compilation.
5.12.9. Intel PenrynUse this property to optimize for Intel Penryn architecture and compatible processors.
x64: Setting this property to Yes adds the -tp=penryn-64 switch to compilation.
5.12.10. Intel Pentium 4Use this property to optimize for Intel Pentium 4 and compatible processors.
5.12.11. Intel Sandy BridgeUse this property to optimize for Intel Sandy Bridge architecture and compatibleprocessors.
PVF Properties
PVF Reference Guide Version 2017 | 129
x64: Setting this property to Yes adds the -tp=sandybridge-64 switch tocompilation.
5.12.12. Generic x86-64 [x64 only]Use this property to optimize for any x86-64 processor-based system.
x64: Setting this property to Yes adds the -tp=px-64 switch to compilation.
5.13. Fortran | Target AcceleratorsThe following properties are available from the Fortran | Target Accelerators propertypage.
For more information about the PGI’s accelerator compilers or on the options in thissection, refer to the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
5.13.1. Target NVIDIA TeslaUse this property to select NVIDIA Tesla targets.
Setting this property to Yes adds the -ta=tesla switch to the PVF compilation andlink lines and activates access to these additional properties:
Tesla Register LimitTesla Use Fused Multiply-AddsTesla Use Fast Math LibraryTesla LLVMTesla NoattachTesla Pin Host MemoryTesla AutocollapseTesla DebugTesla LineinfoTesla UnrollTesla RequiredTesla Flush to ZeroTesla CUDA ToolkitTesla Compute CapabilityTesla: Keep Kernel Files
Important If you change the value of this property and the displayed properties donot change, be sure to click Apply in the property page dialog box.
5.13.2. Tesla Register LimitUse this property to specify the number of registers to use on the GPU.
PVF Properties
PVF Reference Guide Version 2017 | 130
Setting this property to an integer value, n, adds the -ta=tesla:maxregcount:nswitch to the PVF compilation and link lines.
Leaving this property blank indicates no limit to the number of registers to use on theGPU.
5.13.3. Tesla Use Fused Multiply-AddsWhen Target NVIDIA Tesla is set to Yes, use this property to control the generation offused multiply-add (FMA) instructions.
‣ Default: Allows the compiler to control generation of FMA instructions. Thisselection adds no additional sub-options to -ta=tesla.
‣ Yes: Enables generation of FMA instructions by adding -ta=tesla:fma to the PVFcompilation and link lines.
‣ No: Disables generation of FMA instructions by adding -ta=tesla:nofma to thePVF compilation and link lines.
5.13.4. Tesla Use Fast Math LibraryWhen Target NVIDIA Tesla is set to Yes, use this property to use routines from the fastmath library.
Setting this property to Yes adds the -ta=tesla:fastmath switch to the PVFcompilation and link lines.
5.13.5. Tesla LLVMWhen Target NVIDIA Tesla is set to Yes, use this property to control using of LLVMback end.
Setting this property to Yes adds the -ta=tesla:llvm switch to the PVF compilationand link lines.
5.13.6. Tesla NoattachWhen Target NVIDIA Tesla is set to Yes, use this property to prevent attaching toexisting CUDA context.
Setting this property to Yes adds the -ta=tesla:noattach switch to the PVFcompilation and link lines.
5.13.7. Tesla Pin Host MemoryWhen Target NVIDIA Tesla is set to Yes, use this property to set default to pin hostmemory.
Setting this property to Yes adds the -ta=tesla:pin switch to the PVF compilationand link lines.
PVF Properties
PVF Reference Guide Version 2017 | 131
5.13.8. Tesla AutocollapseWhen Target NVIDIA Tesla is set to Yes, use this property to automatically collapsetightly nested loops.
‣ Default: Allows the compiler to control automatic collapse of tightly nested loops.This select adds no additional sub-options to -ta=tesla.
‣ Yes: Enables automatic collapse of tightly nested loops by adding the-ta=tesla:autocollapse switch to the PVF compilation and link lines.
‣ No: Disables automatic collapse of tightly nested loops by adding the-ta=tesla:noautocollapse switch to the PVF compilation and link lines.
5.13.9. Tesla DebugWhen Target NVIDIA Tesla is set to Yes, use this property to control generation of GPUdebug information.
‣ Default: Allows the compiler to control generation of GPU debug information. Thisselect adds no additional sub-options to -ta=tesla.
‣ Yes: Enables generation of GPU debug information by adding the-ta=tesla:debug switch to the PVF compilation and link lines.
‣ No: Disables generation of GPU debug information by adding the-ta=tesla:nodebug switch to the PVF compilation and link lines.
5.13.10. Tesla LineinfoWhen Target NVIDIA Tesla is set to Yes, use this property to control generation of GPUline information.
‣ Default: Allows the compiler to control generation of GPU line information. Thisselect adds no additional sub-options to -ta=tesla.
‣ Yes: Enables generation of GPU line information by adding the-ta=tesla:lineinfo switch to the PVF compilation and link lines.
‣ No: Disables generation of GPU line information by adding the-ta=tesla:nolineinfo switch to the PVF compilation and link lines.
5.13.11. Tesla UnrollWhen Target NVIDIA Tesla is set to Yes, use this property to control automatic innerloop unrolling.
‣ Default: Allows the compiler to control automatic inner loop unrolling. This selectadds no additional sub-options to -ta=tesla.
‣ Yes: Enables automatic inner loop unrolling by adding the -ta=tesla:unrollswitch to the PVF compilation and link lines.
‣ No: Disables automatic inner loop unrolling by adding the -ta=tesla:nounrollswitch to the PVF compilation and link lines.
PVF Properties
PVF Reference Guide Version 2017 | 132
5.13.12. Tesla RequiredWhen Target NVIDIA Tesla is set to Yes, use this property to direct the compiler to issueerror if the compute regions fail to accelerate.
‣ Default: Uses the compiler defaults for handling instances where compute regionscannot be accelerated. This select adds no additional sub-options to -ta=tesla.
‣ Yes: Directs the compiler to stop compilation with an error when it cannot acceleratea compute region by adding the -ta=tesla:required switch to the PVFcompilation and link lines.
‣ No: Directs the compiler to issue warnings when it cannot accelerate a computeregion by adding the -ta=tesla:norequired switch to the PVF compilation andlink lines. Compilation does not stop but accelerator kernels are not generated.
5.13.13. Tesla Flush to ZeroWhen Target NVIDIA Tesla is set to Yes, use this property to control flush-to-zero modefor floating point computations in the GPU code generated for PGI Accelerator modelcompute regions.
‣ Default: Accepts the default handling of floating point computations in the GPUcode generated for CUDA Fortran kernels.
‣ Yes: Enables flush-to-zero mode by adding the -ta=tesla:flushz switch to thePVF compilation and link lines.
‣ No: Disables flush-to-zero mode by adding the -ta=tesla:noflushz switch tothe PVF compilation and link lines.
5.13.14. Tesla Generate RDCWhen Target NVIDIA Tesla is set to Yes, use this property to control generation ofrelocatable device code.
‣ Default: Accepts the compiler’s default generation of relocatable device code.‣ Yes: Directs the compiler to generate relocatable device code by adding
-ta=tesla:rdc switch to the PVF compilation and link lines.‣ No: Prevents the compiler from generating relocatable device code by adding
-ta=tesla:nordc switch to the PVF compilation and link lines.
5.13.15. Tesla CUDA ToolkitWhen Target NVIDIA Tesla is set to Yes, use this property to specify the version of theNVIDIA CUDA toolkit that is targeted by the compilers:
‣ Default: The compiler selects the default CUDA toolkit version.‣ 7.5: Use the default version 7.5 of the CUDA toolkit. This selection adds the
-ta=tesla:cuda7.5 switch to the PVF compilation and link lines.
PVF Properties
PVF Reference Guide Version 2017 | 133
‣ 8.0: Use version 8.0 of the CUDA toolkit. This selection adds the-ta=tesla:cuda8.0 switch to the PVF compilation and link lines.
pgaccelinfo prints the driver version as the first line of output.
For a 7.5 driver: CUDA Driver Version 7050For an 8.0 driver: CUDA Driver Version 8000
5.13.16. Tesla Compute CapabilityWhen Target NVIDIA Tesla is set to Yes, use this property to either automaticallygenerate code compatible with all applicable compute capabilities, or to direct thecompiler to use a manually-selected set.
Select either Automatic or Manual.
‣ Automatic: Let the compiler generate code for all applicable compute capabilities.This is the default.
‣ Manual: Choose one or more compute capabilities to target. The compiler generatescode for each capability specified.
If you select Manual, then you can select any or all of the following computecapabilities that are described in the next sections.
Tesla CC FermiTesla CC Fermi+Tesla CC KeplerTesla CC Kepler+
Important If you select Manual and the additional properties do not appear, clickApply in the Property page dialog.
5.13.17. Tesla CC FermiWhen Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual,use this property to generate code for the Fermi Architecture.
Setting this property to Yes adds the -ta=tesla:fermi switch to the PVF compilationand link lines.
5.13.18. Tesla CC Fermi+When Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual,use this property to generate code for Fermi Architecture and above.
Setting this property to Yes adds the -ta=tesla:fermi+ switch to the PVFcompilation and link lines.
PVF Properties
PVF Reference Guide Version 2017 | 134
5.13.19. Tesla CC KeplerWhen Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual,use this property to generate code for Kepler Architecture.
Setting this property to Yes adds the -ta=tesla:kepler switch to the PVFcompilation and link lines.
5.13.20. Tesla CC Kepler+When Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual,use this property to generate code for Kepler Architecture and above
Setting this property to Yes adds the -ta=tesla:cc30 switch to the PVF compilationand link lines.
5.13.21. Tesla: Keep Kernel FilesWhen Target NVIDIA Tesla is set to Yes, use this property to keep kernel files.
Setting this property to Yes adds the -ta=tesla:keep switch to the PVF compilationand link lines.
5.14. Fortran | DiagnosticsThe following properties are available from the Fortran | Diagnostics property page.These properties allow you to add switches to the compilation line that control theamount and type of information that the compiler provides.
For more information on the options referenced in these pages, refer to MiscellaneousControls and specifically to the -Minfo option.
5.14.1. Warning LevelUse this property to select the level of diagnostic reporting you want the compiler to use.
There are several levels of the -Minform option available through this property. Formore information on this option, refer to Miscellaneous Controls.
5.14.2. Generate AssemblyUse this property to generate an assembly file for each compiled source file.
Setting this property to Yes adds the -Mkeepasm switch to the compilation line.
For more information on -Mkeepasm, refer to Miscellaneous Controls.
PVF Properties
PVF Reference Guide Version 2017 | 135
5.14.3. Annotate AssemblyUse this property to generate assembly files and to annotate the assembly with sourcecode.
Setting this property to Yes adds the -Manno switch to the compilation line.
For more information on -Manno, refer to Miscellaneous Controls.
5.14.4. Accelerator InformationUse this property to generate information about accelerator regions.
Setting this property to Yes adds the -Minfo=accel switch to the compilation line.
5.14.5. CCFF InformationUse this property to append common compiler feedback format (CCFF) information toobject files.
Setting this property to Yes adds the -Minfo=ccff switch to the compilation line.
5.14.6. Fortran Language InformationUse this property to generate information about Fortran language features.
Setting this property to Yes adds the -Minfo=ftn switch to the compilation line.
5.14.7. Inlining InformationUse this property to generate information about inlining.
Setting this property to Yes adds the -Minfo=inline switch to the compilation line.
5.14.8. IPA InformationUse this property to generate information about interprocedural analysis (IPA)optimizations.
Setting this property to Yes adds the -Minfo=ipa switch to the compilation line.
5.14.9. Loop Intensity InformationUse this property to generate compute intensity information about loops.
Setting this property to Yes adds the -Minfo=intensity switch to the compilationline.
5.14.10. Loop Optimization InformationUse this property to generate information about loop optimizations.
PVF Properties
PVF Reference Guide Version 2017 | 136
Setting this property to Yes adds the -Minfo=loop switch to the compilation line.
5.14.11. LRE InformationUse this property to generate information about loop-carried redundancy (LRE)elimination.
Setting this property to Yes adds the -Minfo=lre switch to the compilation line.
5.14.12. OpenMP InformationUse this property to generate information about OpenMP.
Setting this property to Yes adds the -Minfo=mp switch to the compilation line.
5.14.13. Optimization InformationUse this property to generate information about general optimizations.
Setting this property to Yes adds the -Minfo=opt switch to the compilation line.
5.14.14. Parallelization InformationUse this property to generate information about parallel optimizations.
Setting this property to Yes adds the -Minfo=par switch to the compilation line.
5.14.15. Unified Binary InformationUse this property to generate information specific to the PGI Unified Binary.
Setting this property to Yes adds the -Minfo=unified switch to the compilation line.
5.14.16. Vectorization InformationUse this property to generate vectorization information.
Setting this property to Yes adds the -Minfo=vect switch to the compilation line.
5.15. Fortran | ProfilingThe following properties are available from the Fortran | Profiling property page. Theseproperties allow you to add switches to the compilation line that control the informationthat the compiler provides.
For more information on the options referenced in these pages, refer to MiscellaneousControls and specifically to the -Mprof option.
5.15.1. Suppress CCFF InformationUse this property to suppress profiling's default generation of CCFF information.
PVF Properties
PVF Reference Guide Version 2017 | 137
Setting this property to Yes adds the -Mprof=noccff switch to the compiling andlinking lines.
5.15.2. Enable Limited DWARFUse this property to generate limited DWARF information which can be used withperformance profilers.
Setting this property to Yes adds the -Mprof=dwarf switch to the compiling andlinking lines.
5.16. Fortran | RuntimeThe following properties are available from the Fortran | Runtime property page toallow the application to make additional checks at runtime.
5.16.1. Check Array BoundsUse this property to enable array bounds checking at runtime.
Setting this property to Yes adds the -Mbounds switch to the compilation line.
Setting this property to No adds no option to the compilation line, and there is no arraybounds checking at runtime. No is the default.
5.16.2. Check PointersUse this property to perform runtime checks for pointers that are dereferenced whileinitialized to null.
Setting this property to Yes adds the -Mchkptr switch to the compilation line.
Setting this property to No adds no option to the compilation line, and there is noruntime check for pointers that are dereferenced while initialized to null. No is thedefault.
5.16.3. Check StackUse this property to perform runtime stack checks for available space in the prologue ofa function and before the start of a parallel region.
Setting this property to Yes adds the -Mchkstk switch to the compilation line.
Setting this property to No adds no option to the compilation line, and there are noruntime stack checks. No is the default.
5.16.4. Command LineThis property page contains two boxes.
PVF Properties
PVF Reference Guide Version 2017 | 138
‣ The first box, titled All options, is a read-only description of what the compilationline will be. This description is based on the values of the properties set in theFortran property pages.
‣ The second box, titled Additional options, allows you to specify any other optionsthat you want the compiler to use. Use this box when the option you need is notavailable through any of the Fortran property pages.
For more information on all the compiler options that are available, refer toCommand-Line Options Reference.
5.17. Fortran | Command LineThe following properties are available from the Fortran | Command Line property page.
5.17.1. Command LineThis property page contains two boxes.
‣ The first box, titled All options, is a read-only description of what the compilationline will be. This description is based on the values of the properties set in theFortran property pages.
‣ The second box, titled Additional options, allows you to specify any other optionsthat you want the compiler to use. Use this box when the option you need is notavailable through any of the Fortran property pages.
For more information on all the compiler options that are available, refer toCommand-Line Options Reference.
5.18. Linker Property PagesThis section contains the property pages that are included in the Linker category. TheLinker property page category is available for projects that build an executable or adynamically linked library (DLL).
5.19. Linker | GeneralThe following properties are available from the Linker | General property page.
5.19.1. Output FileUse this property to override the default output file name.
Providing the file name and the file’s extension is equivalent to using the -o switch.
You must provide the file’s extension.
For more information on -o, refer to -o.
PVF Properties
PVF Reference Guide Version 2017 | 139
5.19.2. Additional Library DirectoriesUse this property to add one or more directories to the library search path.
For every directory path that is added to this property, PVF adds /LIBPATH:[dir] tothe link line.
There are two ways to add directories to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each directory.‣ Click the ellipsis (‘...’) button in the property page box to open the Additional Library
Directories dialog box.
Enter each directory on its own line in this box. Do not use semi-colons to separatedirectories; the semi-colons are added automatically when the box is closed.
Tip To add directories, use this property. To add libraries, use the AdditionalDependencies property on the Linker | Input Property page.
5.19.3. Stack Reserve SizeUse this property to specify the total number of bytes for stack allocation in virtualmemory. Use decimal notation. This property is equivalent to the -stack=reserveoption. Leave this property blank to direct the linker to choose a default size for thestack.
5.19.4. Stack Commit SizeUse this property to specify the total number of bytes for stack allocation inphysical memory. Use decimal notation. This property is equivalent to the-stack=reserve,commit option. Commit Size is used only if a size is also specified forStack Reserve.
5.19.5. Export SymbolsUse this property to specify whether the DLL will export symbols. This property is onlyvisible for DLL project types.
5.20. Linker | InputThe following properties are available from the Linker | Input property page.
5.20.1. Additional DependenciesUse this property to specify additional dependencies, such as libraries, to the link line.
There are two ways to add libraries to this property:
PVF Properties
PVF Reference Guide Version 2017 | 140
‣ Type the information directly into the property page box.
Use spaces, not semi-colons, to separate multiple libraries. If the name of alibrary contains a space, use double quotes around that library name.
‣ Click the ellipsis (‘...’) button in the property page box to open the AdditionalDependencies dialog box.
Enter each library on its own line in this box.
If you enter two libraries on the same line in this box, PVF interprets these as asingle library whose name contains spaces.
Tip When you close this dialog box, review the contents of the property to make surethat any spaces or double quotes automatically added by PVF are appropriate for yourproject.
5.21. Linker | Command LineThe following properties are available from the Linker | Command Line property page.
5.21.1. Command LineThis property page contains two boxes.
‣ The first box, titled All options, is a read-only description of what the link line will be.This value is based on the values of the properties set in the Linker property pages.
‣ The second box, titled Additional options allows you to specify options that you wantthe linker to use. Use this box when the option you need is not available through anyof the Linker property pages.
For more information on all the compiler options that are available, refer toCommand-Line Options Reference.
5.22. Librarian Property PagesThis section contains the property pages that are included in the Librarian category. TheLibrarian property pages are available for projects that build static libraries.
5.23. Librarian | GeneralThe following properties are available from the Librarian | General property page.
5.23.1. Output FileUse this property to override the default output file name.
PVF Properties
PVF Reference Guide Version 2017 | 141
Providing the file name and the file’s extension is equivalent to using the -o switch.
You must provide the file’s extension.
For more information on -o, refer to -o.
5.23.2. Additional Library DirectoriesUse this property to add one or more directories to the library search path.
For every directory path that is added to this property, PVF adds /LIBPATH:<dir> tothe link line.
There are two ways to add directories to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each directory.‣ Click the ellipsis (‘...’) button in the property page box to open the Additional Library
Directories dialog box.
Enter each directory on its own line in this box. Do not use semi-colons to separatedirectories; the semi-colons are added automatically when the box is closed.
Tip To add directories, use this property. To add libraries, use the AdditionalDependencies property.
5.23.3. Additional DependenciesUse this property to specify additional dependencies, such as libraries, to the link line.
There are two ways to add libraries to this property:
‣ Type the information directly into the property page box.
Use spaces, not semi-colons, to separate multiple libraries. If the name of alibrary contains a space, use double quotes around that library name.
‣ Click the ellipsis (‘...’) button in the property page box to open the AdditionalDependencies dialog box.
Enter each library on its own line in this box.
If you enter two libraries on the same line in this box, PVF interprets these as asingle library whose name contains spaces.
Tip When you close this dialog box, review the contents of the property to make surethat any spaces or double quotes automatically added by PVF are appropriate for yourproject.
PVF Properties
PVF Reference Guide Version 2017 | 142
5.24. Librarian | Command LineThe following properties are available from the Librarian | Command Line propertypage.
5.24.1. Command LineThis property page contains two boxes.
‣ The first box, titled All options, is a read-only description of what the link line willbe. This value is based on the values of the properties set in the Librarian propertypages.
‣ The second box, titled Additional options, allows you to specify options that you wantthe librarian to use, even though these options are not available through any of theLibrarian property pages.
For more information on all the compiler options that are available, refer toCommand-Line Options Reference.
5.25. Resources Property PageThis section contains the property pages that are included in the Resources category.
5.26. Resources | Command LineThe following properties are available from the Resources | Command Line propertypage.
5.26.1. Command LineUse this property to add options to the Resource compiler’s command line.
PVF’s support of resources is somewhat limited at this time and the property pages inthis category reflect that. To add options to the Resource compiler’s command line, usethe Additional options box on this property page.
5.27. Build Events Property PageThis section contains the property pages that are included in the Build Events category.Build events include three types of events: Pre-Build, Pre-Link, and Post-Build.
The Build Events property pages provide an opportunity to specify actions, in additionto compiling and linking, that you want to have happen during the process of a build.
PVF Properties
PVF Reference Guide Version 2017 | 143
5.27.1. Build EventThe name of the build event describes when the event will be fired.
‣ The Pre-Build Event is run before a build starts.‣ The Pre-Link Event is run after compilation but before linking.‣ The Post-Build Event is run after the build completes.
Build events will not be run if a project is up-to-date.
The properties for a build event are the same for all three types of build events.
5.27.2. Command LineUse this property to specify the command line that the build tool will run.
This property is at the core of the build event. For example, to add a time stamp to abuild, you could use time /t as the build event’s command line.
5.27.3. DescriptionUse this property to provide feedback to the Output window. The contents of theDescription property is echoed to the Output window when this event is fired.
5.27.4. Excluded From BuildUse this property to specify whether this build event should be excluded from the buildfor the current configuration.
5.28. Custom Build Step Property PageThis section contains the property pages that are included in the Custom Build Stepcategory.
You can define a custom build step either for a project or for an individual file. Custombuild steps can only be defined for files that are not Fortran or resource files.
5.28.1. Custom Build Step | GeneralThe following properties are available from the Custom Build Step | General propertypage.
5.28.2. Command LineUse this property to specify the command line that the build tool will run. This propertyis at the core of the custom build step.
PVF Properties
PVF Reference Guide Version 2017 | 144
5.28.3. DescriptionUse this property to provide feedback to the Output window. The contents of theDescription property is echoed to the Output window when the custom build step runs.
5.28.4. OutputsUse this property to specify the files generated by the custom build step.
Use semi-colons (‘;’) to separate multiple output files.
When a custom build step is specified at the file-level, this property must be non-empty or the custom build step will be skipped.
5.28.5. Additional DependenciesUse this property to specify any additional input files to use for the custom build.
The custom build step is run when an additional dependency is out of date.
There are two ways to add files to this property:
‣ Type the information directly into the property page box.
Use a semi-colon (‘;’) to separate each directory.‣ Click the ellipsis (‘...’) button in the property page box to open the Additional
Dependencies dialog box.
Enter each file on its own line in this box. Do not use semi-colons to separatedirectories; the semi-colons are added automatically when the box is closed.
PVF Reference Guide Version 2017 | 145
Chapter 6.PVF BUILD MACROS
PVF implements a subset of the build macros supported by Visual C++ along with a fewPVF-specific macros. The macro names are not case-sensitive, and they should be usablein any string field in a property page. Unless otherwise noted, macros that evaluate todirectory names end with a trailing backslash ('\').
In general these items can only be changed if there is an associated PVF project or fileproperty. For example, $(VCInstallDir) cannot be changed, while $(IntDir) canbe changed by modifying the General | Intermediate Directory property.
Table 19 lists the build macros that PVF supports:
Table 19 PVF Build Macros
Macro Name Description
$(Configuration) The name of the current project configuration (for example, "Debug").
$(ConfigurationName) The name of the current project configuration (for example, "Debug").
$(ConfigurationType) The type of the current project configuration - one of the following:
‘Application’‘StaticLibrary’‘DynamicLibrary’
$(DevEnvDir) The installation directory of Visual Studio.
$(InputDir) The directory of the input file. If the project is the input, then this macro isequivalent to $(ProjectDir).
$(InputExt) The file extension of the input file, including the ‘.’ before the fileextension. If the project is the input, then this macro is equivalent to$(ProjectExt).
$(InputFileName) The file name of the input file. If the project is the input, then this macrois equivalent to $(ProjectFileName).
$(InputName) The base name of the input file. If the project is the input, then this macrois equivalent to $(ProjectName).
$(InputPath) The full path name of the input file. If the project is the input, then thismacro is equivalent to $(ProjectPath).
PVF Build Macros
PVF Reference Guide Version 2017 | 146
Macro Name Description
$(IntDir) The path to the directory for intermediate files, relative to the projectdirectory, as set by the Intermediate Directory property.
$(OpenToolsDir) [PVF only]. The location of the Open Tools installation directory, includingfiles needed for building Microsoft Windows applications for 64-bitenvironments.
$(OutDir) The path to the directory for output files, relative to the project directory,as set by the Output Directory property.
$(OutputPath) The path to the directory for output files, relative to the project directory,as set by the Output Directory property.
$(OutputType) The type of the current project output - one of the following:
‘exe’‘staticlibrary’‘library’
$(PGIToolsDir) [PVF only]. The location of the active PGI toolset for 64-bit targets.This directory is the parent of bin, lib, and include directoriescontaining executables, libraries, and include files for the PGI developmentenvironment.
$(Platform) The name of the current project platform (for example, "x64").
$(PlatformArchitecture) The name of the current project platform architecture.
For x64: 64
$(PlatformName) The name of the current project platform (for example, "x64").
$(PlatformShortName) The description of the architecture ABI for the current project platform.
For x64: amd64
$(ProjectDir) The directory of the project.
$(ProjectExt) The file extension of the project file, including the ‘.’ before the fileextension.
$(ProjectFileName) The file name of the project file.
$(ProjectName) The base name of the project.
$(ProjectPath) The full path name of the project.
$(SolutionDir) The directory of the solution.
$(SolutionExt) The file extension of the solution file, including the ‘.’ before the fileextension.
$(SolutionFileName) The file name of the solution file.
$(SolutionName) The base name of the solution.
$(SolutionPath) The full path name of the solution.
$(TargetDir) The directory of the primary output file of the build.
$(TargetExt) The file extension of the primary output file of the build, including the ‘.’before the file extension.
$(TargetFileName) The file name of the primary output file of the build.
PVF Build Macros
PVF Reference Guide Version 2017 | 147
Macro Name Description
$(TargetPath) The full path name of the primary output file of the build.
$(VCInstallDir) The Visual C++ installation directory. If Visual C++ is not installed, thismacro may evaluate to a directory that does not exist.
$(VSInstallDir) The Visual Studio installation directory.
$(WinSDKDir) [PVF only] The location of the Windows SDK installation directory, includingfiles needed for building Microsoft Windows applications for 64-bitenvironments.
PVF Reference Guide Version 2017 | 148
Chapter 7.FORTRAN MODULE/LIBRARYINTERFACES FOR WINDOWS
PGI Visual Fortran provides access to a number of libraries that export C interfaces byusing Fortran modules. PVF uses this mechanism to support the Win32 API and Unix/Linux portability libraries. This section describes the Fortran module library interfacesthat PVF supports, describing each property available.
7.1. Source FilesAll routines described in this section have their prototypes and interfaces described insource files that are included in the PGI Windows compiler installation. The location ofthese files depends on your operating system version and the PGI release version thatyou have installed. These files are typically located in this directory:
C:/Program Files/PGI/win64/[release_version]/src
For example, if you have installed the x64 version of the 17.7 release, look for your filesin this location:
C:/Program Files/PGI/win64/17.7/src
7.2. Data TypesBecause the Win32 API and Portability interfaces resolve to C language libraries, it isimportant to understand how the data types compare within the two languages. Here isa table summarizing how C types correspond with Fortran types for some of the morecommon data types:
Table 20 Fortran Data Type Mappings
Windows Data Type Fortran Data Type
BOOL LOGICAL(4)
BYTE BYTE
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 149
Windows Data Type Fortran Data Type
CHAR CHARACTER
SHORT, WORD INTEGER(2)
DWORD, INT, LONG INTEGER(4)
LONG LONG INTEGER(8)
FLOAT REAL(4)
DOUBLE REAL(8)
x64 Pointers INTEGER(8)
For more information on data types, refer to Fortran Data Types.
7.3. Using DFLIB, LIBM, and DFPORTPVF includes Fortran module interfaces to libraries supporting some standard Clibrary, C math library, and Unix/Linux system call functionality. These functions areprovided by the DFLIB, LIBM, and DFPORT modules. To utilize these modules, add theappropriate USE statement:use dflib
use libm
use dfport
7.3.1. DFLIBTable 21 lists the functions that DFLIB includes. In the table [Generic] refers to ageneric routine. To view the prototype and interfaces, look in the location described inSource Files.
Table 21 DFLIB Function Summary
Routine Result Description
commitqq LOGICAL*4 Executes any pending write operations for the file associated withthe specified unit to the file’s physical device.
delfilesqq INTEGER*4 Deletes the specified files in a specified directory.
findfileqq INTEGER*4 Searches for a file in the directories specified in the PATHenvironment variable.
fullpathqq INTEGER*4 Returns the full path for a specified file or directory.
getdat INTEGER*2,*4,*8 [Generic] Returns the date.
getdrivedirqq INTEGER*4 Returns the current drive and directory path.
getenvqq INTEGER*4 Returns a value from the current environment.
getfileinfoqq INTEGER*4 Returns information about files with names that match thespecified string.
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 150
Routine Result Description
getfileinfoqqi8 INTEGER*4 Returns information about files with names that match thespecified string.
gettim INTEGER*2,*4,*8 [Generic] Returns the time.
makedirqq INTEGER*4 Creates a new directory.
packtimeqq INTEGER*4 Packs the time and date values for use by setfiletimeqq
renamefileqq LOGICAL*4 Renames the specified file.
runqq INTEGER*2 Calls another program and waits for it to execute.
setenvqq LOGICAL*4 Sets the values of an existing environment variable or adds a newone.
setfileaccessqq LOGICAL*4 Sets the file access mode for the specified file.
setfiletimeqq LOGICAL*4 Sets the modification time for the specified file.
signalqq INTEGER*8 Controls signal handling.
sleepqq None Delays execution of the program for a specified time.
splitpathqq LOGICAL*4 Breaks a full path into components.
systemqq LOGICAL*4 Executes a command by passing a command string to theoperating system’s command interpreter.
unpacktimeqq MultipleINTEGERS
Unpacks a file’s packed time and date value into its componentparts.
7.3.2. LIBMA Fortran module called libm is available to declare interfaces to many of the routinesin the standard C math library.Table 22 lists the LIBM routines that are available. To viewthe prototype and interfaces, look in the location described in Source Files.
Some libm routine names conflict with Fortran intrinsics. These routines are not listedin this table because they resolve to Fortran intrinsics.
asin acos atan2 cos cosh
exp log log10 sin sinh
sqrt tan tanh
You can also use libm routines in CUDA Fortran global and device subprograms, inCUF kernels, and in OpenACC compute regions. When targeting NVIDIA devices, thelibm routines translate to the corresponding libm device routine.
Table 22 LIBM Functions
acosf erfc frexp log1p remquo
acosh erff frexpf log1pf remquof
acoshf erfcf ilog log2 rint
asinf expf ilogbf log2f rintf
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 151
asinh exp10 ldexp logb scalbn
asinhf exp10f ldexpf logbf scalbnf
atan2f exp2 lgamma logf scalbln
atanh exp2f lgammaf modf scalblnf
atanhf expf llrint modff sinf
cbrt expm1 llrintf nearbyint sinhf
cbrtf expm1f lrint nearbyintf sqrtf
ceil floor lrint nextafter tanf
ceilf floorf llround nextafterf tanhf
copysign fma llroundf pow tgamma
copysignf fmaf lround powf tgammaf
cosf fmax lroundf remainder trunc
coshf fmaxf log10f remainderf truncf
erf fminf
7.3.3. DFPORTTable 23 lists the functions that DFPORT includes. In the table [Generic] refers to ageneric routine. To view the prototype and interfaces, look in the location described inSource Files.
Table 23 DFPORT Functions
Routine Result Description
abort None Immediately terminates the program. If the operating systemsupports a core dump, abort produces one that can be usedfor debugging.
access INTEGER*4 Determines access mode or existence of a file.
alarm INTEGER*4 Executes a routine after a specified time.
besj0 REAL*4 Computes the BESSEL function of the first kind of order 0 of X,where X is real.
besj1 REAL*4 Computes the BESSEL function of the first kind of order 1 of X,where X is real.
besjn REAL*4 Computes the BESSEL function of the first kind of order N of X,where N is an integer and X is real.
besy0 REAL*4 Computes the BESSEL function of the second kind of order 0 ofX, where X is real.
besy1 REAL*4 Computes the BESSEL function of the second kind of order 1 ofX, where X is real.
besyn REAL*4 Computes the BESSEL function of the second kind of order N ofX, where N is an integer and X is real.
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 152
Routine Result Description
chdir INTEGER*4 Changes the current directory to the directory specified.Returns 0 if successful.
chmod INTEGER*4 Changes the mode of a file by setting the access permissionsof the specified file to the specified mode. Returns 0 ifsuccessful.
ctime STRING(24) Converts and returns the specified time and date as a string.
date STRING Returns the date as a character string: dd-mm-yy.
dbesj0 REAL*8 Computes the double-precision BESSEL function of the firstkind of order 0 of X, where X is a double-precision argument.
dbesj1 REAL*8 Computes the double-precision BESSEL function of the firstkind of order 1 of X, where X is a double-precision argument.
dbesjn REAL*8 Computes the double-precision BESSEL function of the firstkind of order N of X, where N is an integer and X is a double-precision argument.
dbesy0 REAL*8 Computes the double-precision BESSEL function of the secondkind of order 0 of X, where X, where X is a double-precisionargument.
dbesy1 REAL*8 Computes the double-precision BESSEL function of the secondkind of order 1 of X, where X, where X is a double-precisionargument.
dbesyn REAL*8 Computes the double-precision BESSEL function of the secondkind of order N of X, where N is an integer and X, where X is adouble-precision argument.
derf REAL*8 Computes the double-precision error function of X, where X isa double-precision argument.
derfc REAL*8 Computes the complementary double-precision error functionof X, where X is a double-precision argument.
dffrac REAL*8 Returns fractional accuracy of a REAL*8 floating-point value.
dflmax REAL*8 Returns the maximum positive REAL*8 floating-point value.
dflmin REAL*8 Returns the minimum positive REAL*8 floating-point value.
drandm REAL*8 Generates a REAL*8 random number.
dsecnds REAL*8 Returns the number of real time seconds since midnight minusthe supplied argument value.
dtime REAL*4 Returns the elapsed user and system time in seconds since thelast call to dtime.
erf REAL*4 Computes the error function of X, where X is Real.
erfc REAL Computes the complementary error function of X, where X isReal.
etime REAL*4 Returns the elapsed time in seconds since the start of programexecution.
exit None Immediately terminates the program and passes a status tothe parent process.
fdate STRING Returns the current date and time as an ASCII string.
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 153
Routine Result Description
ffrac REAL*4 Returns the fractional accuracy of a REAL*4 floating-pointvalue.
fgetc INTEGER*4 Gets a character or word from an input stream. Returns thenext byte or and integer
flmax REAL*4 Returns the maximum positive REAL*4 lue.
flush None Writes the output to a logical unit.
fputc INTEGER*4 Writes a character or word from an input stream to a logicalunit. Returns 0 if successful or an error.
free None Frees memory previously allocated by MALLOC(). Intended forusers compiling legacy code. Use DEALLOCATE for newer code.
fseek INTEGER*4 Repositions the file pointer associated with the specified file.Returns 0 if successful, 1 otherwise.
fseek64 INTEGER*4 Repositions the file pointer associated with the specifiedstream. Returns 0 if successful, 1 otherwise.
fstat INTEGER*4 Returns file status information about the referenced open fileor shared memory object.
fstat64 INTEGER*4 Returns information in a 64-bit structure about the referencedopen file or shared memory object.
ftell INTEGER*4 Returns the current value of the file pointer associated withthe specified stream.
ftell64 INTEGER*8 Returns the current value of the file pointer associated withthe specified stream.
gerror STRING Writes system error messages.
getarg STRING Returns the list of parameters that were passed to the currentprocess when it was started.
getc INTEGER*4 Retrieves the character at the front of the specified characterlist, or -1 if empty
getcwd INTEGER*4 Retrieves the pathname of the current working directory ornull if fails.
getenv Returns the value of the specified environment variable(s).
getfd INTEGER*4 Returns the file descriptor associated with a Fortran logicalunit.
getgid INTEGER*4 Returns the numerical group ID of the curreni process.
getlog STRING Stores the user’s login name in NAME. If the login name is notfound, then NAME is filled with blanks.
getpid INTEGER*4 Returns the process numerical identifier of the currentprocess.
getuid INTEGER*4 Returns the numerical user ID of the current process.
gmtime INTEGER*4 Converts and returns the date and time formats to GM(Greenwich) time as month, day, and so on.
iargc INTEGER*4 Returns an integer representing the number of arguments forthe last program entered on the command line.
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 154
Routine Result Description
idate INTEGER*4 Returns the date in numerical form, day, month, year.
ierrno INTEGER*4 Returns the system error number for the last error.
inmax INTEGER*4 Returns the maximum positive integer value.
ioinit None Establishes the properties of file I/O for files opened after thecall to ioinit, such as whether to recognize carriage control,how to treat blanks and zeros, and whether to open files atthe beginning or end of the file.
irand1 INTEGER*4 Generates pseudo-random integer in the range of 0 through(2**31)-1, or (2**15)-1 if called with no argument.
irand2 INTEGER*4 Generates pseudo-random integer in the range of 0 through(2**31)-1, or (2**15)-1 if called with no argument.
irandm INTEGER*4 Generates pseudo-random integer in the range of 0 through(2**31)-1, or (2**15)-1 if called with no argument.
isatty LOGICAL Finds the name of a terminal port. Returns TRUE if thespecified unit is a terminal.
itime numerical formof time
Fills and returns TARRAY with numerical values at the currentlocal time, with elements 1,2,and 3 of TARRY being the hour(1-24), minute (1-60) and seconds (1-60).
kill INTEGER*4 Sends the specified signal to the specified process or group ofprocesses. Returns 0 if successful, -1 otherwise
link INTEGER*4 Creates an additional directory entry for the specified existingfile.
lnblnk INTEGER*4 Returns the position of the last non-blank string character inthe specified string.
loc INTEGER*4 Returns the address of an object.
long INTEGER*4 Converts INTEGER*2 to INTEGER*4
lstat INTEGER*4 Obtains information about the referenced open file orshared memory object in a large-file enables programmingenvironment.
lstat64 INTEGER*4 Obtains information in a 64-bit structure about the referencedopen file or shared memory object in a large-file enablesprogramming environment.
ltime Array ofINTEGER*4
Converts the system time from seconds into TARRAY, whichcontains GMT for the current local time zone.
malloc INTEGER*8 Allocates SIZE byes of dynamic memory, returning the addressof the allocated memory. Intended for users compiling legacycode. Use ALLOCATE for newer code.
mclock INTEGER*4 Returns time accounting information about the currentprocess and its child processes in 1/100 or second unitsof measure. The returned value is the sum of the currentprocess’s user time and system time of all child processes.
outstr INTEGER*4 Outputs the value of the specified character to the standardoutput file.
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 155
Routine Result Description
perror None Writes a message to standard error output that describes thelast error encountered by a system call or library subroutine.
putc INTEGER*4 Puts the specified character at the end of the character list.
putenv INTEGER*4 Sets the value of the specified environment variable orcreates a new environment variable.
qsort INTEGER*4 Uses quick-sort algorithm to sort a table of data.
rand1 REAL*4 Provides a method for generating a random number that canbe used as the starting point for the rand procedure.
rand2 REAL*4 Provides a random value between 0 and 1, which is generatedusing the specified seed value, and computed for eachreturned row when used in the select list.
random REAL*4 Uses a non-linear additive feedback random-number generatorto return pseudo-random numbers in the range of 0 to (231-1)
rename INTEGER*4 Renames the specified directory or file
rindex INTEGER*4 Returns the index of the last occurrence of a specific string ofcharacters in a specified string.
rtc REAL*8 Returns the real-time clock value expressed as a number ofclock ticks.
secnds REAL*4 Gets the time in seconds from the real-time system clock. Ifthe value is zero, the time in seconds from midnight is used.
short INTEGER*2 Converts INTEGER*4 to INTEGER*2.
signal INTEGER*4 Specifies the action to take upon delivery of a signal.
sleep None Puts the calling kernel thread to sleep, requiring it to waitfor a wakeup to be issued to continue to run. Provided forcompatibility with older code and should not be used withnew code.
srand1 None Sets the seed for the pseudo-random number generation thatrand1 provides.
srand2 None Sets the seed for the pseudo-random number generation thatrand2 provides.
stat INTEGER*4 Obtains information about the specified file.
stat64 INTEGER*4 Obtains information in a 64-bit structure about the specifiedfile.
stime INTEGER*4 Sets the current value of the specified parameter for thesystem-wide timer.
symlnk INTEGER*4 Creates a symbolic link with the specified name to thespecified file.
system INTEGER*4 Runs a shell command.
time INTEGER*4 Returns the time in seconds since January 1, 1970.
timef REAL*8 Returns the elapsed time in milliseconds since the first call totimef.
times INTEGER*4 Fills the specified structure with time-accounting information.
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 156
Routine Result Description
ttynam STRING(100) Either gets the path name of the terminal or determines if thedevice is a terminal.
unlink INTEGER*4 Removes the specified directory entry, and decreases the linkcount of the file referenced by the link.
wait INTEGER*4 Suspends the calling thread until the process receives a signalthat is not blocked or ignored, or until the calling process’child processes stop or terminate.
7.4. Using the DFWIN moduleThe DFWIN module includes all the modules needed to access the Win32 API. You canuse modules supporting specific portions of the Win32 API separately, but DFWIN isthe only module you need to use the Fortran interfaces to the Win32 API. To use thismodule, add the following line to your Fortran code.use dfwin
To utilize any of the Win32 API interfaces, you can add a Fortran use statement forthe specific library or module that includes it. For example, to use user32.lib, add thefollowing Fortran use statement:use user32
Function calls made through the module interfaces ultimately resolve to C Languageinterfaces, so some accommodation for inter-language calling conventions must be madein the Fortran application. These accommodations include:
‣ On 64-bit platforms, pointers and pointer types such as HANDLE, HINSTANCE,WPARAM, and HWND must be treated as 8-byte quantities (INTEGER(8)).
‣ In general, C makes calls by value while Fortran makes calls by reference.‣ When doing Windows development one must sometimes provide callback functions
for message processing, dialog processing, etc. These routines are called by theWindows system when events are processed. To provide the expected functionsignature for a callback function, the user may need to use the STDCALL attributedirective (!DEC$ ATTRIBUTE::STDCALL) in the declaration.
7.5. Supported Libraries and ModulesThe following tables provide lists of the functions in each library or module that PGIsupports in DFWIN.
For information on the interfaces associated with these functions, refer to the fileslocated here:C:\Program Files\PGI\win64\17.7\src
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 157
7.5.1. advapi32The following table lists the functions that advapi32 includes:
AccessCheckAndAuditAlarm AccessCheckByType
AccessCheckByTypeAndAuditAlarm AccessCheckByTypeResultList
AccessCheckByTypeResultListAndAuditAlarm AccessCheckByTypeResultListAndAuditAlarmByHandle
AddAccessAllowedAce AddAccessAllowedAceEx
AddAccessAllowedObjectAce AddAccessDeniedAce
AddAccessDeniedAceEx AddAccessDeniedObjectAce
AddAce AddAuditAccessAce
AddAuditAccessAceEx AddAuditAccessObjectAce
AdjustTokenGroups AdjustTokenPrivileges
AllocateAndInitializeSid AllocateLocallyUniqueId
AreAllAccessesGranted AreAnyAccessesGranted
BackupEventLog CheckTokenMembership
ClearEventLog CloseEncryptedFileRaw
CloseEventLog ConvertToAutoInheritPrivateObjectSecurity
CopySid CreatePrivateObjectSecurity
CreatePrivateObjectSecurityEx CreatePrivateObjectSecurityWithMultipleInheritance
CreateProcessAsUser CreateProcessWithLogonW
CreateProcessWithTokenW CreateRestrictedToken
CreateWellKnownSid DecryptFile
DeleteAce DeregisterEventSource
DestroyPrivateObjectSecurity DuplicateToken
DuplicateTokenEx EncryptFile
EqualDomainSid EqualPrefixSid
EqualSid FileEncryptionStatus
FindFirstFreeAce FreeSid
GetAce GetAclInformation
GetCurrentHwProfile GetEventLogInformation
GetFileSecurity GetKernelObjectSecurity
GetLengthSid GetNumberOfEventLogRecords
GetOldestEventLogRecord GetPrivateObjectSecurity
GetSecurityDescriptorControl GetSecurityDescriptorDacl
GetSecurityDescriptorGroup GetSecurityDescriptorLength
GetSecurityDescriptorOwner GetSecurityDescriptorRMControl
GetSecurityDescriptorSacl GetSidIdentifierAuthority
GetSidLengthRequired GetSidSubAuthority
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 158
GetSidSubAuthorityCount GetTokenInformation
GetUserName GetWindowsAccountDomainSid
ImpersonateAnonymousToken ImpersonateLoggedOnUser
ImpersonateNamedPipeClient ImpersonateSelf
InitializeAcl InitializeSecurityDescriptor
InitializeSid IsTextUnicode
IsTokenRestricted IsTokenUntrusted
IsValidAcl IsValidSecurityDescriptor
IsValidSid IsWellKnownSid
LogonUser LogonUserEx
LookupAccountName LookupAccountSid
LookupPrivilegeDisplayName LookupPrivilegeName
LookupPrivilegeValue MakeAbsoluteSD
MakeAbsoluteSD2 MakeSelfRelativeSD
MapGenericMask NotifyChangeEventLog
ObjectCloseAuditAlarm ObjectDeleteAuditAlarm
ObjectOpenAuditAlarm ObjectPrivilegeAuditAlarm
OpenBackupEventLog OpenEncryptedFileRaw
OpenEventLog OpenProcessToken
OpenThreadToken PrivilegeCheck
PrivilegedServiceAuditAlarm ReadEncryptedFileRaw
ReadEventLog RegisterEventSource
ReportEvent RevertToSelf
SetAclInformation SetFileSecurity
SetKernelObjectSecurity SetPrivateObjectSecurity
SetPrivateObjectSecurityEx SetSecurityDescriptorControl
SetSecurityDescriptorDacl SetSecurityDescriptorGroup
SetSecurityDescriptorOwner SetSecurityDescriptorRMControl
SetSecurityDescriptorSacl SetThreadToken
SetTokenInformation WriteEncryptedFileRaw
7.5.2. comdlg32The following table lists the functions that comdlg32 includes:
AfxReplaceText ChooseColor ChooseFont
CommDlgExtendedError FindText GetFileTitle
GetOpenFileName GetSaveFileName PageSetupDlg
PrintDlg PrintDlgEx ReplaceText
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 159
7.5.3. dfwbaseThese are the functions that dfwbase includes:
chartoint LoByte MakeWord
chartoreal LoWord MakeWparam
CopyMemory LoWord64 PaletteIndex
GetBlueValue MakeIntAtom PaletteRGB
GetGreenValue MakeIntResource PrimaryLangID
GetRedValue MakeLangID RGB
HiByte MakeLCID RtlCopyMemory
HiWord MakeLong SortIDFromLCID
HiWord64 MakeLParam SubLangID
inttochar MakeLResult
7.5.4. dfwintyThese are the functions that dfwinty includes:
dwNumberOfFunctionKeys rdFunction
7.5.5. gdi32These are the functions that gdi32 includes:
AbortDoc AbortPath AddFontMemResourceEx
AddFontResource AddFontResourceEx AlphaBlend
AngleArc AnimatePalette Arc
ArcTo BeginPath BitBlt
CancelDC CheckColorsInGamut ChoosePixelFormat
Chord CloseEnhMetaFile CloseFigure
CloseMetaFile ColorCorrectPalette ColorMatchToTarget
CombineRgn CombineTransform CopyEnhMetaFile
CopyMetaFile CreateBitmap CreateBitmapIndirect
CreateBrushIndirect CreateColorSpace CreateCompatibleBitmap
CreateCompatibleDC CreateDC CreateDIBitmap
CreateDIBPatternBrush CreateDIBPatternBrushPt CreateDIBSection
CreateDiscardableBitmap CreateEllipticRgn CreateEllipticRgnIndirect
CreateEnhMetaFile CreateFont CreateFontIndirect
CreateFontIndirectEx CreateHalftonePalette CreateHatchBrush
CreateIC CreateMetaFile CreatePalette
CreatePatternBrush CreatePen CreatePenIndirect
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 160
CreatePolygonRgn CreatePolyPolygonRgn CreateRectRgn
CreateRectRgnIndirect CreateRoundRectRgn CreateScalableFontResource
CreateSolidBrush DeleteColorSpace DeleteDC
DeleteEnhMetaFile DeleteMetaFile DeleteObject
DescribePixelFormat DeviceCapabilities DPtoLP
DrawEscape Ellipse EndDoc
EndPage EndPath EnumEnhMetaFile
EnumFontFamilies EnumFontFamiliesEx EnumFonts
EnumICMProfiles EnumMetaFile EnumObjects
EqualRgn Escape ExcludeClipRect
ExtCreatePen ExtCreateRegion ExtEscape
ExtFloodFill ExtSelectClipRgn ExtTextOut
FillPath FillRgn FixBrushOrgEx
FlattenPath FloodFill FrameRgn
GdiComment GdiFlush GdiGetBatchLimit
GdiSetBatchLimit GetArcDirection GetAspectRatioFilterEx
GetBitmapBits GetBitmapDimensionEx GetBkColor
GetBkMode GetBoundsRect GetBrushOrgEx
GetCharABCWidthsA GetCharABCWidthsFloat GetCharABCWidthsI
GetCharABCWidthsW GetCharacterPlacement GetCharWidth
GetCharWidth32 GetCharWidthFloat GetCharWidthI
GetClipBox GetClipRgn GetColorAdjustment
GetColorSpace GetCurrentObject GetCurrentPositionEx
GetDCBrushColor GetDCOrgEx GetDCPenColor
GetDeviceCaps GetDeviceGammaRamp GetDIBColorTable
GetDIBits GetEnhMetaFile GetEnhMetaFileBits
GetEnhMetaFileDescriptionA GetEnhMetaFileDescriptionW GetEnhMetaFileHeader
GetEnhMetaFilePaletteEntries GetEnhMetaFilePixelFormat GetFontData
GetFontLanguageInfo GetFontUnicodeRanges GetGlyphIndices
GetGlyphOutline GetGraphicsMode GetICMProfileA
GetICMProfileW GetKerningPairs GetLayout
GetLogColorSpace GetMapMode GetMetaFile
GetMetaFileBitsEx GetMetaRgn GetMiterLimit
GetNearestColor GetNearestPaletteIndex GetObject
GetObjectType GetOutlineTextMetrics GetPaletteEntries
GetPath GetPixel GetPixelFormat
GetPolyFillMode GetRandomRgn GetRasterizerCaps
GetRegionData GetRgnBox GetROP2
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 161
GetStockObject GetStretchBltMode GetSystemPaletteEntries
GetSystemPaletteUse GetTextAlign GetTextCharacterExtra
GetTextCharset GetTextCharsetInfo GetTextColor
GetTextExtentExPoint GetTextExtentExPointI GetTextExtentPoint
GetTextExtentPoint32 GetTextExtentPointI GetTextFace
GetTextMetrics GetViewportExtEx GetViewportOrgEx
GetWindowExtEx GetWindowOrgEx GetWinMetaFileBits
GetWorldTransform GradientFill IntersectClipRect
InvertRgn LineDD LineTo
LPtoDP MaskBlt ModifyWorldTransform
MoveToEx OffsetClipRgn OffsetRgn
OffsetViewportOrgEx OffsetWindowOrgEx PaintRgn
PatBlt PathToRegion Pie
PlayEnhMetaFile PlayEnhMetaFileRecord PlayMetaFile
PlayMetaFileRecord PlgBlt PolyBezier
PolyBezierTo PolyDraw Polygon
Polyline PolylineTo PolyPolygon
PolyPolyline PolyTextOut PtInRegion
PtVisible RealizePalette Rectangle
RectInRegion RectVisible RemoveFontMemResourceEx
RemoveFontResource RemoveFontResourceEx ResetDC
ResizePalette RestoreDC RoundRect
SaveDC ScaleViewportExtEx ScaleWindowExtEx
SelectClipPath SelectClipRgn SelectObject
SelectPalette SetAbortProc SetArcDirection
SetBitmapBits SetBitmapDimensionEx SetBkColor
SetBkMode SetBoundsRect SetBrushOrgEx
SetColorAdjustment SetColorSpace SetDCBrushColor
SetDCPenColor SetDeviceGammaRamp SetDIBColorTable
SetDIBits SetDIBitsToDevice SetEnhMetaFileBits
SetGraphicsMode SetICMMode SetICMProfile
SetLayout SetMapMode SetMapperFlags
SetMetaFileBitsEx SetMetaRgn SetMiterLimit
SetPaletteEntries SetPixel SetPixelFormat
SetPixelV SetPolyFillMode SetRectRgn
SetROP2 SetStretchBltMode SetSystemPaletteUse
SetTextAlign SetTextCharacterExtra SetTextColor
SetTextJustification SetViewportExtEx SetViewportOrgEx
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 162
SetWindowExtEx SetWindowOrgEx SetWinMetaFileBits
SetWorldTransform StartDoc StartPage
StretchBlt StretchDIBits StrokeAndFillPath
SwapBuffers TextOut
TranslateCharsetInfo TransparentBlt UnrealizeObject
UpdateColors UpdateICMRegKey wglCopyContext
wglCreateContext wglCreateLayerContext wglDeleteContext
wglDescribeLayerPlane wglGetCurrentContext wglGetCurrentDC
wglGetLayerPaletteEntries wglGetProcAddress wglMakeCurrent
wglRealizeLayerPalette wglSetLayerPaletteEntries wglShareLists
wglSwapLayerBuffers wglSwapMultipleBuffers wglUseFontBitmaps
wglUseFontOutlines WidenPath
7.5.6. kernel32These are the functions that kernel32 includes:
ActivateActCtx AddAtom
AddConsoleAlias AddRefActCtx
AddVectoredContinueHandler AddVectoredExceptionHandler
AllocateUserPhysicalPages AllocConsole
AreFileApisANSI AssignProcessToJobObject
AttachConsole BackupRead
BackupSeek BackupWrite
Beep BeginUpdateResource
BindIoCompletionCallback BuildCommDCB
BuildCommDCBAndTimeouts CallNamedPipe
CancelDeviceWakeupRequest CancelIo
CancelTimerQueueTimer CancelWaitableTimer
CheckNameLegalDOS8Dot3 CheckRemoteDebuggerPresent
ClearCommBreak ClearCommError
CloseHandle CommConfigDialog
CompareFileTime ConnectNamedPipe
ContinueDebugEvent ConvertFiberToThread
ConvertThreadToFiber ConvertThreadToFiberEx
CopyFile CopyFileEx
CreateActCtx CreateConsoleScreenBuffer
CreateDirectory CreateDirectoryEx
CreateEvent CreateFiber
CreateFiberEx CreateFile
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 163
CreateFileMapping CreateHardLink
CreateIoCompletionPort CreateJobObject
CreateJobSet CreateMailslot
CreateMemoryResourceNotification CreateMutex
CreateNamedPipe CreatePipe
CreateProcess CreateRemoteThread
CreateSemaphore CreateTapePartition
CreateThread CreateTimerQueue
CreateTimerQueueTimer CreateWaitableTimer
DeactivateActCtx DebugActiveProcess
DebugActiveProcessStop DebugBreak
DebugBreakProcess DebugSetProcessKillOnExit
DecodePointer DecodeSystemPointer
DefineDosDevice DeleteAtom
DeleteCriticalSection DeleteFiber
DeleteFile DeleteTimerQueue
DeleteTimerQueueEx DeleteTimerQueueTimer
DeleteVolumeMountPoint DeviceIoControl
DisableThreadLibraryCalls DisconnectNamedPipe
DnsHostnameToComputerName DosDateTimeToFileTime
DuplicateHandle EncodePointer
EncodeSystemPointer EndUpdateResource
EnterCriticalSection EnumResourceLanguages
EnumResourceNames EnumResourceTypes
EnumSystemFirmwareTables EraseTape
EscapeCommFunction ExitProcess
ExitThread ExpandEnvironmentStrings
FatalAppExit FatalExit
FileTimeToDosDateTime FileTimeToLocalFileTime
FileTimeToSystemTime FillConsoleOutputAttribute
FillConsoleOutputCharacter FindActCtxSectionGuid
FindActCtxSectionString FindAtom
FindClose FindCloseChangeNotification
FindFirstChangeNotification FindFirstFile
FindFirstFileEx FindFirstVolume
FindFirstVolumeMountPoint FindNextChangeNotification
FindNextFile FindNextVolume
FindNextVolumeMountPoint FindResource
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 164
FindResourceEx FindVolumeClose
FindVolumeMountPointClose FlsAlloc
FlsFree FlsGetValue
FlsSetValue FlushConsoleInputBuffer
FlushFileBuffers FlushInstructionCache
FlushViewOfFile FormatMessage
FreeConsole FreeEnvironmentStrings
FreeLibrary FreeLibraryAndExitThread
FreeResource FreeUserPhysicalPages
GenerateConsoleCtrlEvent GetAtomName
GetBinaryType GetCommandLine
GetCommConfig GetCommMask
GetCommModemStatus GetCommProperties
GetCommState GetCommTimeouts
GetCompressedFileSize GetComputerName
GetConsoleAlias GetConsoleAliases
GetConsoleAliasesLength GetConsoleAliasExes
GetConsoleAliasExesLength GetConsoleCP
GetConsoleCursorInfo GetConsoleDisplayMode
GetConsoleFontSize GetConsoleMode
GetConsoleOutputCP GetConsoleProcessList
GetConsoleScreenBufferInfo GetConsoleSelectionInfo
GetConsoleTitle GetConsoleWindow
GetCurrentActCtx GetCurrentConsoleFont
GetCurrentDirectory GetCurrentProcess
GetCurrentProcessId GetCurrentProcessorNumber
GetCurrentThread GetCurrentThreadId
GetDefaultCommConfig GetDevicePowerState
GetDiskFreeSpace GetDiskFreeSpaceEx
GetDllDirectory GetDriveType
GetEnvironmentStrings GetEnvironmentVariable
GetExitCodeProcess GetExitCodeThread
GetFileAttributes GetFileAttributesEx
GetFileInformationByHandle GetFileSize
GetFileSizeEx GetFileTime
GetFileType GetFirmwareEnvironmentVariable
GetFullPathName GetHandleInformation
GetLargePageMinimum GetLargestConsoleWindowSize
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 165
GetLastError GetLocalTime
GetLogicalDrives GetLogicalDriveStrings
GetLogicalProcessorInformation GetLongPathName
GetMailslotInfo GetModuleFileName
GetModuleHandle GetModuleHandleEx
GetNamedPipeHandleState GetNamedPipeInfo
GetNativeSystemInfo GetNumaAvailableMemoryNode
GetNumaHighestNodeNumber GetNumaNodeProcessorMask
GetNumaProcessorNode GetNumberOfConsoleInputEvents
GetNumberOfConsoleMouseButtons GetOverlappedResult
GetPriorityClass GetPrivateProfileInt
GetPrivateProfileSection GetPrivateProfileSectionNames
GetPrivateProfileString GetPrivateProfileStruct
GetProcAddress GetProcessAffinityMask
GetProcessHandleCount GetProcessHeap
GetProcessHeaps GetProcessId
GetProcessIdOfThread GetProcessIoCounters
GetProcessPriorityBoost GetProcessShutdownParameters
GetProcessTimes GetProcessVersion
GetProcessWorkingSetSize GetProcessWorkingSetSizeEx
GetProfileInt GetProfileSection
GetProfileString GetQueuedCompletionStatus
GetShortPathName GetVolumeNameForVolumeMountPoint
GetVolumePathName GetVolumePathNamesForVolumeName
GetWindowsDirectory GetWriteWatch
GlobalAddAtom GlobalAlloc
GlobalCompact GlobalDeleteAtom
GlobalFindAtom GlobalFix
GlobalFlags GlobalFree
GlobalGetAtomName GlobalHandle
GlobalLock GlobalMemoryStatus
GlobalMemoryStatusEx GlobalReAlloc
GlobalSize GlobalUnfix
GlobalUnlock GlobalUnWire
GlobalWire HeapAlloc
HeapCompact HeapCreate
HeapDestroy HeapFree
HeapLock HeapQueryInformation
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 166
HeapReAlloc HeapSetInformation
HeapSize HeapUnlock
HeapValidate HeapWalk
InitAtomTable InitializeCriticalSection
InitializeCriticalSectionAndSpinCount InitializeSListHead
InterlockedCompareExchange InterlockedCompareExchange64
InterlockedDecrement InterlockedExchange
InterlockedExchangeAdd InterlockedFlushSList
InterlockedIncrement InterlockedPopEntrySList
InterlockedPushEntrySList IsBadCodePtr
IsBadHugeReadPtr IsBadHugeWritePtr
IsBadReadPtr IsBadStringPtr
IsBadWritePtr IsDebuggerPresent
IsProcessInJob IsProcessorFeaturePresent
IsSystemResumeAutomatic LeaveCriticalSection
LoadLibrary LoadLibraryEx
LoadModule LoadResource
LocalAlloc LocalCompact
LocalFileTimeToFileTime LocalFlags
LocalFree LocalHandle
LocalLock LocalReAlloc
LocalShrink LocalSize
LocalUnlock LockFile
LockFileEx LockResource
lstrcat lstrcmp
lstrcmpi lstrcpy
lstrcpyn lstrlen
MapUserPhysicalPages MapUserPhysicalPagesScatter
MapViewOfFile MapViewOfFileEx
MoveFile MoveFileEx
MoveFileWithProgress MulDiv
NeedCurrentDirectoryForExePath OpenEvent
OpenFile OpenFileMapping
OpenJobObject OpenMutex
OpenProcess OpenSemaphore
OpenThread OpenWaitableTimer
OutputDebugString PeekConsoleInput
PeekNamedPipe PostQueuedCompletionStatus
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 167
PrepareTape ProcessIdToSessionId
PulseEvent PurgeComm
QueryActCtxW QueryDepthSList
QueryDosDevice QueryInformationJobObject
QueryMemoryResourceNotification QueryPerformanceCounter
QueryPerformanceFrequency QueueUserAPC
QueueUserWorkItem RaiseException
ReadConsole ReadConsoleInput
ReadConsoleOutput ReadConsoleOutputAttribute
ReadConsoleOutputCharacter ReadDirectoryChangesW
ReadFile ReadFileEx
ReadFileScatter ReadProcessMemory
RegisterWaitForSingleObject RegisterWaitForSingleObjectEx
ReleaseActCtx ReleaseMutex
ReleaseSemaphore RemoveDirectory
RemoveVectoredContinueHandler RemoveVectoredExceptionHandler
ReOpenFile ReplaceFile
RequestDeviceWakeup RequestWakeupLatency
ResetEvent ResetWriteWatch
RestoreLastError ResumeThread
ScrollConsoleScreenBuffer SearchPath
SetCommBreak SetCommConfig
SetCommMask SetCommState
SetCommTimeouts SetComputerName
SetComputerNameEx SetConsoleActiveScreenBuffer
SetConsoleCP SetConsoleCtrlHandler
SetConsoleCursorInfo SetConsoleCursorPosition
SetConsoleMode SetConsoleOutputCP
SetConsoleScreenBufferSize SetConsoleTextAttribute
SetConsoleTitle SetConsoleWindowInfo
SetCriticalSectionSpinCount SetCurrentDirectory
SetDefaultCommConfig SetDllDirectory
SetEndOfFile SetEnvironmentStrings
SetEnvironmentVariable SetErrorMode
SetEvent SetFileApisToANSI
SetFileApisToOEM SetFileAttributes
SetFilePointer SetFilePointerEx
SetFileShortName SetFileTime
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 168
SetFileValidData SetFirmwareEnvironmentVariable
SetHandleCount SetHandleInformation
SetInformationJobObject SetLastError
SetLocalTime SetMailslotInfo
SetMessageWaitingIndicator SetNamedPipeHandleState
SetPriorityClass SetProcessAffinityMask
SetProcessPriorityBoost SetProcessShutdownParameters
SetProcessWorkingSetSize SetProcessWorkingSetSizeEx
SetStdHandle SetSystemTime
SetSystemTimeAdjustment SetTapeParameters
SetTapePosition SetThreadAffinityMask
SetThreadContext SetThreadExecutionState
SetThreadIdealProcessor SetThreadPriority
SetThreadPriorityBoost SetThreadStackGuarantee
SetTimerQueueTimer SetTimeZoneInformation
SetUnhandledExceptionFilter SetupComm
SetVolumeLabel SetVolumeMountPoint
SetWaitableTimer SignalObjectAndWait
SizeofResource Sleep
SleepEx SuspendThread
SwitchToFiber SwitchToThread
SystemTimeToFileTime SystemTimeToTzSpecificLocalTime
TerminateJobObject TerminateProcess
TerminateThread TlsAlloc
TlsFree TlsGetValue
TlsSetValue TransactNamedPipe
TransmitCommChar TryEnterCriticalSection
TzSpecificLocalTimeToSystemTime UnhandledExceptionFilter
UnlockFile UnlockFileEx
UnmapViewOfFile UnregisterWait
UnregisterWaitEx UpdateResource
VerifyVersionInfo VirtualAlloc
VirtualAllocEx VirtualFree
VirtualFreeEx VirtualLock
VirtualProtect VirtualProtectEx
VirtualQuery VirtualQueryEx
VirtualUnlock WaitCommEvent
WaitForDebugEvent WaitForMultipleObjects
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 169
WaitForMultipleObjectsEx WaitForSingleObject
WaitForSingleObjectEx WaitNamedPipe
WinExec Wow64DisableWow64FsRedirection
Wow64EnableWow64FsRedirection Wow64RevertWow64FsRedirection
WriteConsole WriteConsoleInput
WriteConsoleOutput WriteConsoleOutputAttribute
WriteConsoleOutputCharacter WriteFile
WriteFileEx WriteFileGather
WritePrivateProfileSection WritePrivateProfileString
WritePrivateProfileStruct WriteProcessMemory
WriteProfileSection WriteProfileString
WriteTapemark WTSGetActiveConsoleSessionId
ZombifyActCtx _hread
_hwrite _lclose
_lcreat _llseek
_lopen _lread
_lwrite
7.5.7. shell32These are the functions that shell32 includes:
DoEnvironmentSubst ShellExecuteEx
DragAcceptFiles Shell_NotifyIcon
DragFinish SHEmptyRecycleBin
DragQueryFile SHFileOperation
DragQueryPoint SHFreeNameMappings
DuplicateIcon SHGetDiskFreeSpaceEx
ExtractAssociatedIcon SHGetFileInfo
ExtractIcon SHGetNewLinkInfo
ExtractIconEx SHInvokePrinterCommand
FindExecutable SHIsFileAvailableOffline
IsLFNDrive SHLoadNonloadedIconOverlayIdentifiers
SHAppBarMessage SHQueryRecycleBin
SHCreateProcessAsUserW SHSetLocalizedName
ShellAbout WinExecError
ShellExecute
7.5.8. user32These are the functions that user32 includes:
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 170
ActivateKeyboardLayout AdjustWindowRect AdjustWindowRectEx
AllowSetForegroundWindow AnimateWindow AnyPopup
AppendMenu ArrangeIconicWindows AttachThreadInput
BeginDeferWindowPos BeginPaint BringWindowToTop
BroadcastSystemMessage BroadcastSystemMessageEx CallMsgFilter
CallNextHookEx CallWindowProc CascadeWindows
ChangeClipboardChain ChangeDisplaySettings ChangeDisplaySettingsEx
ChangeMenu CharLower CharLowerBuff
CharNext CharNextEx CharPrev
CharPrevEx CharToOem CharToOemBuff
CharUpper CharUpperBuff CheckDlgButton
CheckMenuItem CheckMenuRadioItem CheckRadioButton
ChildWindowFromPoint ChildWindowFromPointEx ClientToScreen
ClipCursor CloseClipboard CloseDesktop
CloseWindow CloseWindowStation CopyAcceleratorTable
CopyCursor CopyIcon CopyImage
CopyRect CountClipboardFormats CreateAcceleratorTable
CreateCaret CreateCursor CreateDesktop
CreateDialogIndirectParam CreateDialogParam CreateIcon
CreateIconFromResource CreateIconFromResourceEx CreateIconIndirect
CreateMDIWindow CreateMenu CreatePopupMenu
CreateWindow CreateWindowEx CreateWindowStation
DeferWindowPos DefFrameProc DefMDIChildProc
DefRawInputProc DefWindowProc DeleteMenu
DeregisterShellHookWindow DestroyAcceleratorTable DestroyCaret
DestroyCursor DestroyIcon DestroyMenu
DestroyWindow DialogBoxIndirectParam DialogBoxParam1
DialogBoxParam2 DisableProcessWindowsGhosting DispatchMessage
DlgDirList DlgDirListComboBox DlgDirSelectComboBoxEx
DlgDirSelectEx DragDetect DragObject
DrawAnimatedRects DrawCaption DrawEdge
DrawFocusRect DrawFrameControl DrawIcon
DrawIconIndirect DrawMenuBar DrawState
DrawText DrawTextEx EmptyClipboard
EnableMenuItem EnableScrollBar EnableWindow
EndDeferWindowPos EndDialog EndMenu
EndPaint EndTask EnumChildWindows
EnumClipboardFormats EnumDesktops EnumDesktopWindows
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 171
EnumDisplayDevices EnumDisplayMonitors EnumDisplaySettings
EnumDisplaySettingsEx EnumProps EnumPropsEx
EnumThreadWindows EnumWindows EnumWindowStations
EqualRect ExcludeUpdateRgn ExitWindowsEx
FillRect FindWindow FindWindowEx
FlashWindow FlashWindowEx FrameRect
GetActiveWindow GetAltTabInfo GetAncestor
GetAsyncKeyState GetCapture GetCaretBlinkTime
GetCaretPos GetClassInfo GetClassInfoEx
GetClassLong GetClassLongPtr GetClassName
GetClassWord GetClientRect GetClipboardData
GetClipboardFormatName GetClipboardOwner GetClipboardSequenceNumber
GetClipboardViewer GetClipCursor GetComboBoxInfo
GetCursor GetCursorInfo GetCursorPos
GetDC GetDCEx GetDesktopWindow
GetDialogBaseUnits GetDlgCtrlID GetDlgItem
GetDlgItemInt GetDlgItemText GetDoubleClickTime
GetFocus GetForegroundWindow GetGuiResources
GetGUIThreadInfo GetIconInfo GetInputState
GetKBCodePage GetKeyboardLayout GetKeyboardLayoutList
GetKeyboardLayoutName GetKeyboardState GetKeyboardType
GetKeyNameText GetKeyState GetLastActivePopup
GetLastInputInfo GetLayeredWindowAttributes GetListBoxInfo
GetMenu GetMenuBarInfo GetMenuCheckMarkDimensions
GetMenuContextHelpId GetMenuDefaultItem GetMenuInfo
GetMenuItemCount GetMenuItemID GetMenuItemInfo
GetMenuItemRect GetMenuState GetMenuString
GetMessage GetMessageExtraInfo GetMessagePos
GetMessageTime GetMonitorInfo GetMouseMovePointsEx
GetNextDlgGroupItem GetNextDlgTabItem GetOpenClipboardWindow
GetParent GetPriorityClipboardFormat GetProcessDefaultLayout
GetProcessWindowStation GetProp GetQueueStatus
GetRawInputBuffer GetRawInputData GetRawInputDeviceInfo
GetRawInputDeviceList GetRegisteredRawInputDevices GetScrollBarInfo
GetScrollInfo GetScrollPos GetScrollRange
GetShellWindow GetSubMenu GetSysColor
GetSysColorBrush GetSystemMenu GetSystemMetrics
GetTabbedTextExtent GetThreadDesktop GetTitleBarInfo
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 172
GetTopWindow GetUpdateRect GetUpdateRgn
GetUserObjectInformation GetUserObjectSecurity GetWindow
GetWindowContextHelpId GetWindowDC GetWindowInfo
GetWindowLong GetWindowLongPtr GetWindowModuleFileName
GetWindowPlacement GetWindowRect GetWindowRgn
GetWindowRgnBox GetWindowText GetWindowTextLength
GetWindowThreadProcessId GetWindowWord GrayString
HideCaret HiliteMenuItem InflateRect
InSendMessage InSendMessageEx InsertMenu
InsertMenuItem InternalGetWindowText IntersectRect
InvalidateRect InvalidateRgn InvertRect
IsCharAlpha IsCharAlphaNumeric IsCharLower
IsCharUpper IsChild IsClipboardFormatAvailable
IsDialogMessage IsDlgButtonChecked IsGUIThread
IsHungAppWindow IsIconic IsMenu
IsRectEmpty IsWindow IsWindowEnabled
IsWindowUnicode IsWindowVisible IsWinEventHookInstalled
IsWow64Message IsZoomed keybd_event
KillTimer LoadAccelerators LoadBitmap
LoadCursor1 LoadCursor2 LoadCursorFromFile
LoadIcon1 LoadIcon2 LoadImage
LoadKeyboardLayout LoadMenu1 LoadMenu2
LoadMenuIndirect LoadString LockSetForegroundWindow
LockWindowUpdate LockWorkStation LookupIconIdFromDirectory
LookupIconIdFromDirectoryEx LRESULT MapDialogRect
MapVirtualKey MapVirtualKeyEx MapWindowPoints
MenuItemFromPoint MessageBeep MessageBox
MessageBoxEx MessageBoxIndirect ModifyMenu1
ModifyMenu2 MonitorFromPoint MonitorFromRect
MonitorFromWindow mouse_event MoveWindow
MsgWaitForMultipleObjects MsgWaitForMultipleObjectsEx NotifyWinEvent
OemKeyScan OemToChar OemToCharBuff
OffsetRect OpenClipboard OpenDesktop
OpenIcon OpenInputDesktop OpenWindowStation
PaintDesktop PeekMessage PostMessage
PostQuitMessage PostThreadMessage PrintWindow
PrivateExtractIcons PtInRect RealChildWindowFromPoint
RealGetWindowClass RedrawWindow RegisterClass
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 173
RegisterClassEx RegisterClipboardFormat RegisterDeviceNotification
RegisterHotKey RegisterRawInputDevices RegisterShellHookWindow
RegisterWindowMessage ReleaseCapture ReleaseDC
RemoveMenu RemoveProp ReplyMessage
ScreenToClient ScrollDC ScrollWindow
ScrollWindowEx SendDlgItemMessage SendInput
SendMessage SendMessageCallback SendMessageTimeout
SendNotifyMessage SetActiveWindow SetCapture
SetCaretBlinkTime SetCaretPos SetClassLong
SetClassLongPtr SetClassWord SetClipboardData
SetClipboardViewer SetCursor SetCursorPos
SetDebugErrorLevel SetDlgItemInt SetDlgItemText
SetDoubleClickTime SetFocus SetForegroundWindow
SetKeyboardState SetLastErrorEx SetLayeredWindowAttributes
SetMenu SetMenuContextHelpId SetMenuDefaultItem
SetMenuInfo SetMenuItemBitmaps SetMenuItemInfo
SetMessageExtraInfo SetMessageQueue SetParent
SetProcessDefaultLayout SetProcessWindowStation SetProp
SetRect SetRectEmpty SetScrollInfo
SetScrollPos SetScrollRange SetSysColors
SetSystemCursor SetThreadDesktop SetTimer
SetUserObjectInformation SetUserObjectSecurity SetWindowContextHelpId
SetWindowLong SetWindowLongPtr SetWindowPlacement
SetWindowPos SetWindowRgn SetWindowsHook
SetWindowsHookEx SetWindowText SetWindowWord
SetWinEventHook ShowCaret ShowCursor
ShowOwnedPopups ShowScrollBar ShowWindow
ShowWindowAsync SubtractRect SwapMouseButton
SwitchDesktop SwitchToThisWindow SystemParametersInfo
TabbedTextOut TileWindows ToAscii
ToAsciiEx ToUnicode ToUnicodeEx
TrackMouseEvent TrackPopupMenu TrackPopupMenuEx
TranslateAccelerator TranslateMDISysAccel TranslateMessage
UnhookWindowsHook UnhookWindowsHookEx UnhookWinEvent
UnionRect UnloadKeyboardLayout UnregisterClass
UnregisterDeviceNotification UnregisterHotKey UpdateLayeredWindow
UpdateLayeredWindowIndirect UpdateWindow UserHandleGrantAccess
ValidateRect ValidateRgn VkKeyScan
Fortran Module/Library Interfaces for Windows
PVF Reference Guide Version 2017 | 174
VkKeyScanEx WaitForInputIdle WaitMessage
WindowFromDC WindowFromPoint WinHelp
wsprintf wvsprintf
7.5.9. winverThese are the functions that winver includes:
GetFileVersionInfo VerFindFile VerLanguageName
GetFileVersionInfoSize VerInstallFile VerQueryValue
7.5.10. wsock32These are the functions that wsock32 includes:
accept AcceptEx bind
closesocket connect GetAcceptExSockaddrs
getpeername gethostname getprotobyname
getprotobynumber getservbyname getservbyport
getsockname getsockopt htonl
htons inet_addr inet_ntoa
ioctlsocket listen ntohl
ntohs recv select
send sendto setsockopt
shutdown socket TransmitFile
WSAAsyncGetHostByName WSAAsyncGetProtoByName WSAAsyncGetProtoByNumber
WSAAsyncGetServByName WSAAsyncGetServByPort WSAAsyncSelect
WSACancelAsyncRequest WSACancelBlockingCall WSACleanup
WSAGetLastError WSAIsBlocking WSARecvEx
WSASetBlockingHook WSASetLastError WSAStartup
PVF Reference Guide Version 2017 | 175
Chapter 8.MESSAGES
This section describes the various messages that the compiler produces. These messagesinclude the sign-on message and diagnostic messages for remarks, warnings, and errors.The compiler always displays any error messages, along with the erroneous source line,on the screen. If you specify the -Mlist option, the compiler places any error messagesin the listing file. You can also use the -v option to display more information aboutthe compiler, assembler, and linker invocations and about the host system. For moreinformation on the -Mlist and -v options, refer to ‘Using Command-line Options’ inthe PGI Compiler User’s Guide.
8.1. Diagnostic MessagesDiagnostic messages provide syntactic and semantic information about your source text.Syntactic information includes information such as syntax errors. Semantic informationincludes information such as unreachable code, incorrect number of arguments specifiedfor a call to a routine, illegal data type usage, etc.
You can specify that the compiler displays error messages at a certain level with the -Minform option.
The compiler messages refer to a severity level, a message number, and the line numberwhere the error occurs.
The compiler can also display internal error messages on standard error. If yourcompilation produces any internal errors, please contact the PGI technical reportingservice, href="https://www.pgicompilers.com/support/support_request.php.
If you use the listing file option -Mlist, the compiler places diagnostic messages afterthe source lines in the listing file, in the following format: PGFTN-etype-enum-message (filename: line)
Where:etype
is a character signifying the severity levelenum
is the error number
Messages
PVF Reference Guide Version 2017 | 176
messageis the error message
filenameis the source filename
lineis the line number where the compiler detected an error.
8.2. Phase Invocation MessagesYou can display compiler, assembler, and linker phase invocations by using the -vcommand line option. For further information about this option, refer to the ‘UsingCommand-line Options’ section of the PVF User's Guide, https://www.pgroup.com/resources/docs.php.
8.3. Fortran Compiler Error MessagesThis section presents the error messages generated by the PGF77, PGF95, andPGFORTRAN compilers. The compilers display error messages in the program listingand on standard output. They can also display internal error messages on standarderror.
8.3.1. Message FormatEach message is numbered. Each message also lists the line and column number wherethe error occurs. A dollar sign ($) in a message represents information that is specific toeach occurrence of the message.
8.3.2. Message ListError message severities:I
informativeW
warningS
severe errorF
fatal errorV
variableV000 Internal compiler error. $ $
This message indicates an error in the compiler, rather than a user error – although itmay be possible for a user error to cause an internal error. The severity may vary; if it isinformative or warning, correct object code was probably generated, but it is not safe torely on this. Regardless of the severity or cause, internal errors should be reported to thePGI technical reporting service, https://www.pgroup.com/support/support_request.php.
Messages
PVF Reference Guide Version 2017 | 177
F001 Source input file name not specified
On the command line, source file name should be specified either before all the switches,or after them.F002 Unable to open source input file: $
Source file name is misspelled, file is not in current working directory, or file is readprotected.F003 Unable to open listing file
This message typically occurs when the user does not have write permission for thecurrent working directory.F004 $ $
Generic message for file errors.F005 Unable to open temporary file
Compiler uses directory specified by the environment variables $TMP or $TMPDIR inwhich to create temporary files. If neither of these directories is available on the node onwhich the compiler is being used, this error will occur.S006 Input file empty
Source input file does not contain any Fortran statements other than comments orcompiler directives.F007 Subprogram too large to compile at this optimization level $
Internal compiler data structure overflow, working storage exhausted, or some othernon-recoverable problem related to the size of the subprogram. If this error occursat opt level 2, reducing the opt level to 1 may work around the problem. Movingthe subprogram being compiled to its own source file may eliminate the problem. Ifthis error occurs while compiling a subprogram of fewer than 2000, please report theproblem to the PGI technical reporting service, https://www.pgroup.com/support/support_request.php.F008 Error limit exceeded
The compiler gives up because too many severe errors were issued; the error limit can bereset on the command line.F009 Unable to open assembly file
This message typically occurs when the user does not have write permission for thecurrent working directory.F010 File write error occurred $
The file system may be full.S011 Unrecognized command line switch: $
Refer to the PGI Compiler User’s Guide for a list of allowed compiler switches.S012 Value required for command line switch: $
Certain switches require an immediately following value, such as "-opt 2".S013 Unrecognized value specified for command line switch: $
Messages
PVF Reference Guide Version 2017 | 178
S014 Ambiguous command line switch: $
Too short an abbreviation was used for one of the switches.W015 Hexadecimal or octal constant truncated to fit data type
I016 Identifier, $, truncated to 63 chars
An identifier may be at most 63 characters in length; characters after the 63rd areignored.S017 Unable to open include file: $
File is missing, read protected, or maximum include depth (10) exceeded. Rememberthat the file name should be enclosed in quotes.S018 Illegal label $ $
Used for label ‘field’ errors or illegal values. E.g., in fixed source form, the label field(first five characters) of the indicated line contains a non-numeric character.S019 Illegally placed continuation line
A continuation line does not follow an initial line, or more than 99 continuation lineswere specified.S020 Unrecognized compiler directive
Refer to Directives Reference for list of allowed compiler directives.S021 Label field of continuation line is not blank
The first five characters of a continuation line must be blank.S022 Unexpected end of file - missing END statement
The source file is missing and END statement, or the file is truncated.S023 Syntax error - unbalanced $
Unbalanced parentheses or brackets.W024 CHARACTER or Hollerith constant truncated to fit data type
A character or hollerith constant was converted to a data type that was not large enoughto contain all of the characters in the constant. This type conversion occurs when theconstant is used in an arithmetic expression or is assigned to a non-character variable.The character or hollerith constant is truncated on the right, that is, if 4 characters areneeded then the first 4 are used and the remaining characters are discarded.W025 Illegal character ($) - ignored
The current line contains a character, possibly non-printing, which is not a legal Fortrancharacter (characters inside of character or Hollerith constants cannot cause this error).As a general rule, all non-printing characters are treated as white space characters(blanks and tabs); no error message is generated when this occurs. If for some reason, anon-printing character is not treated as a white space character, its hex representation isprinted in the form dd where each d is a hex digit.S026 Unmatched quote
A character constant is missing a closing quote or the source file is truncated.S027 Illegal integer constant: $
Messages
PVF Reference Guide Version 2017 | 179
Integer constant is too large for 32 bit word.S028 Illegal real or double precision constant: $
S029 Illegal $ constant: $
Illegal hexadecimal, octal, or binary constant. A hexadecimal constant consists of digits0..9 and letters A..F or a..f; any other character in a hexadecimal constant is illegal. Anoctal constant consists of digits 0..7; any other digit or character in an octal constant isillegal. A binary constant consists of digits 0 or 1; any other digit or character in a binaryconstant is illegal.S030 Explicit shape must be specified for $
A shape for an array expression is effected in this context.S031 Illegal data type length specifier for $
The data type length specifier (e.g. 4 in INTEGER*4) is not a constant expression that is amember of the set of allowed values for this particular data type.W032 Data type length specifier not allowed for $
The data type length specifier (e.g. 4 in INTEGER*4) is not allowed in the given syntax(e.g. DIMENSION A(10)*4).S033 Illegal use of constant $
A constant was used in an illegal context, such as on the left side of an assignmentstatement or as the target of a data initialization statement.S034 Syntax error at or near $
Illegal command specified.I035 Predefined intrinsic $ loses intrinsic property
An intrinsic name was used in a manner inconsistent with the language definition forthat intrinsic. The compiler, based on the context, will treat the name as a variable or anexternal function.S036 Illegal implicit character range
First character must alphabetically precede second.S037 Contradictory data type specified for $
The indicated identifier appears in more than one type specification statement anddifferent data types are specified for it.S038 Symbol, $, has not been explicitly declared
The indicated identifier must be declared in a type statement; this is required when theIMPLICIT NONE statement occurs in the subprogram.W039 Symbol, $, appears illegally in a SAVE statement $
An identifier appearing in a SAVE statement must be a local variable or array.S040 Illegal common variable $
Indicated identifier is a dummy variable, is already in a common block, or haspreviously been defined to be something other than a variable or array.W041 Illegal use of dummy argument $
Messages
PVF Reference Guide Version 2017 | 180
This error can occur in several situations. It can occur if dummy arguments werespecified on a PROGRAM statement. It can also occur if a dummy argument nameoccurs in a DATA, COMMON, SAVE, or EQUIVALENCE statement. A programstatement must have an empty argument list.S042 $ is a duplicate dummy argument
Each dummy argument must have a unique name.S043 Illegal attempt to redefine $ $
An attempt was made to define a symbol in a manner inconsistent with an earlierdefinition of the same symbol. This can happen for a number of reasons. The messageattempts to indicate the situation that occurred.
intrinsic – An attempt was made to redefine an intrinsic function. A symbol thatrepresents an intrinsic function may be redefined if that symbol has not been previouslyverified to be an intrinsic function. For example, the intrinsic sin can be defined to bean integer array. If a symbol is verified to be an intrinsic function via the INTRINSICstatement or via an intrinsic function reference then it must be referred to as an intrinsicfunction for the remainder of the program unit.
symbol – An attempt was made to redefine a symbol that was previously defined.An example of this is to declare a symbol to be a PARAMETER which was previouslydeclared to be a subprogram argument.S044 Multiple declaration for symbol $
A redundant declaration of a symbol has occurred. For example, an attempt was madeto declare a symbol as an ENTRY when that symbol was previously declared as anENTRY.S045 Data type of entry point $ disagrees with function $
The current function has entry points with data types inconsistent with the data type ofthe current function. For example, the function returns type character and an entry pointreturns type complex.S046 Data type length specifier in wrong position
The CHARACTER data type specifier has a different position for the length specifierfrom the other data types. Suppose, we want to declare arrays ARRAYA and ARRAYBto have 8 elements each having an element length of 4 bytes. The difference is thatARRAYA is character and ARRAYB is integer. The declarations would be CHARACTERARRAYA(8)*4 and INTEGER ARRAYB*4(8).S047 More than seven dimensions specified for array
The compiler currently supports up to seven dimensions for arrays.S048 Illegal use of '*' in declaration of array $
An asterisk may be used only as the upper bound of the last dimension.S049 Illegal use of '*' in non-subroutine subprogram
The alternate return specifier ‘*’ is legal only in the subroutine statement. Programs,functions, and block data are not allowed to have alternate return specifiers.S050 Assumed size array, $, is not a dummy argument
Messages
PVF Reference Guide Version 2017 | 181
Arrays with ‘*’ in their dimension(s) may only be declared as dummy arguments.S051 Unrecognized built-in % function
The allowable built-in functions are %VAL, %REF, %LOC, and %FILL. One wasencountered that did not match one of these allowed forms.S052 Illegal argument to %VAL or %LOC
S053 %REF or %VAL not legal in this context
The built-in functions %REF and %VAL can only be used as actual parameters inprocedure calls.W054 Implicit character $ used in a previous implicit statement
An implicit character has been given an implied data type more than once. The implieddata type for the implicit character is changed anyway.W055 Multiple implicit none statements
The IMPLICIT NONE statement can occur only once in a subprogram.W056 Implicit type declaration
The -Mdclchk switch and an implicit declaration following an IMPLICIT NONEstatement will produce a warning message for IMPLICIT statements.S057 Illegal equivalence of dummy variable, $
Dummy arguments may not appear in EQUIVALENCE statements.S058 Equivalenced variables $ and $ not in same common block
A common block variable must not be equivalenced with a variable in another commonblock.S059 Conflicting equivalence between $ and $
The indicated equivalence implies a storage layout inconsistent with other equivalences.S060 Illegal equivalence of structure variable, $
STRUCTURE and UNION variables may not appear in EQUIVALENCE statements.S061 Equivalence of $ and $ extends common block backwards
W062 Equivalence forces $ to be unaligned
EQUIVALENCE statements have defined an address for the variable which has analignment not optimal for variables of its data type. This can occur when INTEGER andCHARACTER data are equivalenced, for instance.I063 Gap in common block $ before $
S064 Illegal use of $ in DATA statement implied DO loop
The indicated variable is referenced where it is not an active implied DO index variable.S065 Repeat factor less than zero
S066 Too few data constants in initialization statement
S067 Too many data constants in initialization statement
Messages
PVF Reference Guide Version 2017 | 182
S068 Numeric initializer for CHARACTER $ out of range 0 through 255
A CHARACTER*1 variable or character array element can be initialized to an integer,octal, or hexadecimal constant if that constant is in the range 0 through 255.S069 Illegal implied DO expression
The only operations allowed within an implied DO expression are integer +, -, *, and /.S070 Incorrect sequence of statements $
The statement order is incorrect. For instance, an IMPLICIT NONE statement mustprecede a specification statement which in turn must precede an executable statement.S071 Executable statements not allowed in block data
S072 Assignment operation illegal to $ $
The destination of an assignment operation must be a variable, array reference, or vectorreference. The assignment operation may be by way of an assignment statement, a datastatement, or the index variable of an implied DO-loop. The compiler has determinedthat the identifier used as the destination is not a storage location. The error messageattempts to indicate the type of entity used.
entry point – An assignment to an entry point that was not a function procedure wasattempted.
external procedure – An assignment to an external procedure or a Fortran intrinsicname was attempted. If the identifier is the name of an entry point that is not a function,an external procedure.S073 Intrinsic or predeclared, $, cannot be passed as an argument
S074 Illegal number or type of arguments to $ $
The indicated symbol is an intrinsic or generic function, or a predeclared subroutine orfunction, requiring a certain number of arguments of a fixed data type.S075 Subscript, substring, or argument illegal in this context for $
This can happen if you try to doubly index an array such as ra(2)(3). This also applies tosubstring and function references.S076 Subscripts specified for non-array variable $
S077 Subscripts omitted from array $
S078 Wrong number of subscripts specified for $
S079 Keyword form of argument illegal in this context for $$
S080 Subscript for array $ is out of bounds
S081 Illegal selector $ $
S082 Illegal substring expression for variable $
Substring expressions must be of type integer and if constant must be greater than zero.S083 Vector expression used where scalar expression required
Messages
PVF Reference Guide Version 2017 | 183
A vector expression was used in an illegal context. For example, iscalar = iarray,where a scalar is assigned the value of an array. Also, character and record references arenot vectorizable.S084 Illegal use of symbol $ $
This message is used for many different errors.S085 Incorrect number of arguments to statement function $
S086 Dummy argument to statement function must be a variable
S087 Non-constant expression where constant expression required
S088 Recursive subroutine or function call of $
A function may not call itself.S089 Illegal use of symbol, $, with character length = *
Symbols of type CHARACTER*(*) must be dummy variables and must not be used asstatement function dummy parameters and statement function names. Also, a dummyvariable of type CHARACTER*(*) cannot be used as a function.S090 Hollerith constant more than 4 characters
In certain contexts, Hollerith constants may not be more than 4 characters long.S091 Constant expression of wrong data type
S092 Illegal use of variable length character expression
A character expression used as an actual argument, or in certain contexts within I/Ostatements, must not consist of a concatenation involving a passed length charactervariable.W093 Type conversion of expression performed
An expression of some data type appears in a context which requires an expression ofsome other data type. The compiler generates code to convert the expression into therequired type.S094 Variable $ is of wrong data type $
The indicated variable is used in a context which requires a variable of some other datatype.S095 Expression has wrong data type
An expression of some data type appears in a context which requires an expression ofsome other data type.S096 Illegal complex comparison
The relations .LT., .GT., .GE., and .LE. are not allowed for complex values.S097 Statement label $ has been defined more than once
More than one statement with the indicated statement number occurs in thesubprogram.S098 Divide by zero
Messages
PVF Reference Guide Version 2017 | 184
S099 Illegal use of $
Aggregate record references may only appear in aggregate assignment statements,unformatted I/O statements, and as parameters to subprograms. They may not appear,for example, in expressions. Also, records with differing structure types may not beassigned to one another.S100 Expression cannot be promoted to a vector
An expression was used that required a scalar quantity to be promoted to a vectorillegally. For example, the assignment of a character constant string to a character array.Records, too, cannot be promoted to vectors.S101 Vector operation not allowed on $
Record and character typed entities may only be referenced as scalar quantities.S102 Arithmetic IF expression has wrong data type
The parenthetical expression of an arithmetic if statement must be an integer, real, ordouble precision scalar expression.S103 Type conversion of subscript expression for $
The data type of a subscript expression must be integer. If it is not, it is converted.S104 Illegal control structure $
This message is issued for a number of errors involving IF-THEN statements, DO loops,and directives. You may see one of the following messages:PGF90-S-0104-Illegal control structure - unterminated PARALLEL directivePGF90-S-0104-Illegal control structure - unterminated block IF
If the line number specified is the last line (END statement) of the subprogram, the erroris probably an unterminated DO loop or IF-THEN statement. If the message containsunterminated PARALLEL directive, it is likely you are missing the required !$ompend parallel directive.S105 Unmatched ELSEIF, ELSE or ENDIF statement
An ELSEIF, ELSE, or ENDIF statement cannot be matched with a preceding IF-THENstatement.S106 DO index variable must be a scalar variable
The DO index variable cannot be an array name, a subscripted variable, a PARAMETERname, a function name, a structure name, etc.S107 Illegal assigned goto variable $
S108 Illegal variable, $, in NAMELIST group $
A NAMELIST group can only consist of arrays and scalars.I109 Overflow in $ constant $, constant truncated at left
A non-decimal (hexadecimal, octal, or binary) constant requiring more than 64-bitsproduces an overflow. The constant is truncated at left (e.g. ‘1234567890abcdef1’x will be‘234567890abcdef1’x).I110 <reserved message number>
Messages
PVF Reference Guide Version 2017 | 185
I111 Underflow of real or double precision constant
I112 Overflow of real or double precision constant
S113 Label $ is referenced but never defined
S114 Cannot initialize $
W115 Assignment to DO variable $ in loop
S116 Illegal use of pointer-based variable $ $
S117 Statement not allowed within a $ definition
The statement may not appear in a STRUCTURE or derived type definition.S118 Statement not allowed in DO, IF, or WHERE block
I119 Redundant specification for $
Data type of indicated symbol specified more than once.I120 Label $ is defined but never referenced
I121 Operation requires logical or integer data types
An operation in an expression was attempted on data having a data type incompatiblewith the operation. For example, a logical expression can consist of only logical elementsof type integer or logical. Real data would be invalid.I122 Character string truncated
Character string or Hollerith constant appearing in a DATA statement or PARAMETERstatement has been truncated to fit the declared size of the corresponding identifier.W123 Hollerith length specification too big, reduced
The length specifier field of a hollerith constant specified more characters than werepresent in the character field of the hollerith constant. The length specifier was reducedto agree with the number of characters present.S124 Relational expression mixes character with numeric data
A relational expression is used to compare two arithmetic expressions or two characterexpressions. A character expression cannot be compared to an arithmetic expression.I125 Dummy procedure $ not declared EXTERNAL
A dummy argument which is not declared in an EXTERNAL statement is used as thesubprogram name in a CALL statement, or is called as a function, and is thereforeassumed to be a dummy procedure. This message can result from a failure to declare adummy array.I126 Name $ is not an intrinsic function
I127 Optimization level for $ changed to opt 1 $
W128 Integer constant truncated to fit data type: $
An integer constant will be truncated when assigned to data types smaller than 32-bits,such as a BYTE.
Messages
PVF Reference Guide Version 2017 | 186
I129 Floating point overflow. Check constants and constant expressions
I130 Floating point underflow. Check constants and constant expressions
I131 Integer overflow. Check floating point expressions cast to integer
I132 Floating pt. invalid oprnd. Check constants and constant expressions
I133 Divide by 0.0. Check constants and constant expressions
S134 Illegal attribute $ $
W135 Missing STRUCTURE name field
A STRUCTURE name field is required on the outermost structure.W136 Field-namelist not allowed
The field-namelist field of the STRUCTURE statement is disallowed on the outermoststructure.W137 Field-namelist is required in nested structures
W138 Multiply defined STRUCTURE member name $
A member name was used more than once within a structure.W139 Structure $ in RECORD statement not defined
A RECORD statement contains a reference to a STRUCTURE that has not yet beendefined.S140 Variable $ is not a RECORD
S141 RECORD required on left of $
S142 $ is not a member of this RECORD
S143 $ requires initializer
W144 NEED ERROR MESSAGE $ $
This is used as a temporary message for compiler development.W145 %FILL only valid within STRUCTURE block
The %FILL special name was used outside of a STRUCTURE multiline statement. Itis only valid when used within a STRUCTURE multiline statement even though it isignored.S146 Expression must be character type
S147 Character expression not allowed in this context
S148 Reference to $ required
An aggregate reference to a record was expected during statement compilation butanother data type was found instead.S149 Record where arithmetic value required
Messages
PVF Reference Guide Version 2017 | 187
An aggregate record reference was encountered when an arithmetic expression wasexpected.S150 Structure, Record, derived type, or member $ not allowed in this context
A structure, record, or member reference was found in a context which is not supported.S151 Empty TYPE, STRUCTURE, UNION, or MAP
TYPE - ENDTYPE, STRUCTURE - ENDSTRUCTURE, UNION - ENDUNION or MAP -ENDMAP declaration contains no members.S152 All dimension specifiers must be ':'
S153 Array objects are not conformable $
S154 DISTRIBUTE target, $, must be a processor
S155 $ $
S156 Number of colons and triplets must be equal in ALIGN $ with $
S157 Illegal subscript use of ALIGN dummy $ - $
S158 Alternate return not specified in SUBROUTINE or ENTRY
An alternate return can only be used if alternate return specifiers appeared in theSUBROUTINE or ENTRY statements.S159 Alternate return illegal in FUNCTION subprogram
An alternate return cannot be used in a FUNCTION.S160 ENDSTRUCTURE, ENDUNION, or ENDMAP does not match top
S161 Vector subscript must be rank-one array
W162 Not equal test of loop control variable $ replaced with < or > test.
S163 <reserved message number>
S164 Overlapping data initializations of $
An attempt was made to data initialize a variable or array element already initialized.S165 $ appeared more than once as a subprogram
A subprogram name appeared more than once in the source file. The message isapplicable only when an assembly file is the output of the compiler.S166 $ cannot be a common block and a subprogram
A name appeared as a common block name and a subprogram name. The message isapplicable only when an assembly file is the output of the compiler.I167 Inconsistent size of common block $
A common block occurs in more than one subprogram of a source file and its size isnot identical. The maximum size is chosen. The message is applicable only when anassembly file is the output of the compiler.S168 Incompatible size of common block $
Messages
PVF Reference Guide Version 2017 | 188
A common block occurs in more than one subprogram of a source file and is initializedin one subprogram. Its initialized size was found to be less than its size in the othersubprogram(s). The message is applicable only when an assembly file is the output ofthe compiler.W169 Multiple data initializations of common block $
A common block is initialized in more than one subprogram of a source file. Only thefirst set of initializations apply. The message is applicable only when an assembly file isthe output of the compiler.W170 PGI Fortran extension: $ $
Use of a nonstandard feature. A description of the feature is provided.W171 PGI Fortran extension: nonstandard statement type $
W172 PGI Fortran extension: numeric initialization of CHARACTER $
A CHARACTER*1 variable or array element was initialized with a numeric value.W173 PGI Fortran extension: nonstandard use of data type length specifier
W174 PGI Fortran extension: type declaration contains data initialization
W175 PGI Fortran extension: IMPLICIT range contains nonalpha characters
W176 PGI Fortran extension: nonstandard operator $
W177 PGI Fortran extension: nonstandard use of keyword argument $
W178 <reserved message number>
W179 PGI Fortran extension: use of structure field reference $
W180 PGI Fortran extension: nonstandard form of constant
W181 PGI Fortran extension: & alternate return
W182 PGI Fortran extension: mixed non-character and character elements in COMMON$
W183 PGI Fortran extension: mixed non-character and character EQUIVALENCE ($,$)
W184 Mixed type elements (numeric and/or character types) in COMMON $
W185 Mixed numeric and/or character type EQUIVALENCE ($,$)
S186 Argument missing for formal argument $
S187 Too many arguments specified for $
S188 Argument number $ to $: type mismatch
S189 Argument number $ to $: association of scalar actual argument to array dummyargument
S190 Argument number $ to $: non-conformable arrays
Messages
PVF Reference Guide Version 2017 | 189
S191 Argument number $ to $ cannot be an assumed-size array
S192 Argument number $ to $ must be a label
W193 Argument number $ to $ does not match INTENT (OUT)
W194 INTENT(IN) argument cannot be defined - $
S195 Statement may not appear in an INTERFACE block $
S196 Deferred-shape specifiers are required for $
S197 Invalid qualifier or qualifier value (/$) in OPTIONS statement
An illegal qualifier was found or a value was specified for a qualifier which does notexpect a value. In either case, the qualifier for which the error occurred is indicated inthe error message.S198 $ $ in ALLOCATE/DEALLOCATE
W199 Unaligned memory reference
A memory reference occurred whose address does not meet its data alignmentrequirement.S200 Missing UNIT/FILE specifier
S201 Illegal I/O specifier - $
S202 Repeated I/O specifier - $
S203 FORMAT statement has no label
S204 $ $
Miscellaneous I/O error.S205 Illegal specification of scale factor
The integer following + or - has been omitted, or P does not follow the integer value.S206 Repeat count is zero
S207 Integer constant expected in edit descriptor
S208 Period expected in edit descriptor
S209 Illegal edit descriptor
S210 Exponent width not used in the Ew.dEe or Gw.dEe edit descriptors
S211 Internal I/O not allowed in this I/O statement
S212 Illegal NAMELIST I/O
Namelist I/O cannot be performed with internal, unformatted, formatted, and list-directed I/O. Also, I/O lists must not be present.S213 $ is not a NAMELIST group name
Messages
PVF Reference Guide Version 2017 | 190
S214 Input item is not a variable reference
S215 Assumed sized array name cannot be used as an I/O item or specifier
An assumed size array was used as an item to be read or written or as an I/O specifier(i.e., FMT = array-name). In these contexts the size of the array must be known.S216 STRUCTURE/UNION cannot be used as an I/O item
S217 ENCODE/DECODE buffer must be a variable, array, or array element
S218 Statement labeled $ $
S219 <reserved message number>
S220 Redefining predefined macro $
S221 #elif after #else
A preprocessor #elif directive was found after a #else directive; only #endif is allowed inthis context.S222 #else after #else
A preprocessor #else directive was found after a #else directive; only #endif is allowed inthis context.S223 #if-directives too deeply nested
Preprocessor #if directive nesting exceeded the maximum allowed (currently 10).S224 Actual parameters too long for $
The total length of the parameters in a macro call to the indicated macro exceeded themaximum allowed (currently 2048).W225 Argument mismatch for $
The number of arguments supplied in the call to the indicated macro did not agree withthe number of parameters in the macro’s definition.F226 Can't find include file $
The indicated include file could not be opened.S227 Definition too long for $
The length of the macro definition of the indicated macro exceeded the maximumallowed (currently 2048).S228 EOF in comment
The end of a file was encountered while processing a comment.S229 EOF in macro call to $
The end of a file was encountered while processing a call to the indicated macro.S230 EOF in string
The end of a file was encountered while processing a quoted string.S231 Formal parameters too long for $
Messages
PVF Reference Guide Version 2017 | 191
The total length of the parameters in the definition of the indicated macro exceeded themaximum allowed (currently 2048).S232 Identifier too long
The length of an identifier exceeded the maximum allowed (currently 2048).S233 <reserved message number>
W234 Illegal directive name
The sequence of characters following a # sign was not an identifier.W235 Illegal macro name
A macro name was not an identifier.S236 Illegal number $
The indicated number contained a syntax error.F237 Line too long
The input source line length exceeded the maximum allowed (currently 2048).W238 Missing #endif
End of file was encountered before a required #endif directive was found.W239 Missing argument list for $
A call of the indicated macro had no argument list.S240 Number too long
The length of a number exceeded the maximum allowed (currently 2048).W241 Redefinition of symbol $
The indicated macro name was redefined.I242 Redundant definition for symbol $
A definition for the indicated macro name was found that was the same as a previousdefinition.F243 String too long
The length of a quoted string exceeded the maximum allowed (currently 2048).S244 Syntax error in #define, formal $ not identifier
A formal parameter that was not an identifier was used in a macro definition.W245 Syntax error in #define, missing blank after name or arglist
There was no space or tab between a macro name or argument list and the macro’sdefinition.S246 Syntax error in #if
A syntax error was found while parsing the expression following a #if or #elif directive.S247 Syntax error in #include
The #include directive was not correctly formed.W248 Syntax error in #line
Messages
PVF Reference Guide Version 2017 | 192
A #line directive was not correctly formed.W249 Syntax error in #module
A #module directive was not correctly formed.W250 Syntax error in #undef
A #undef directive was not correctly formed.W251 Token after #ifdef must be identifier
The #ifdef directive was not followed by an identifier.W252 Token after #ifndef must be identifier
The #ifndef directive was not followed by an identifier.S253 Too many actual parameters to $
The number of actual arguments to the indicated macro exceeded the maximum allowed(currently 31).S254 Too many formal parameters to $
The number of formal arguments to the indicated macro exceeded the maximumallowed (currently 31).F255 Too much pushback
The preprocessor ran out of space while processing a macro expansion. The macro maybe recursive.W256 Undefined directive $
The identifier following a # was not a directive name.F257 POS value must be positive.
A value for POS <= 0 was encountered. Negative and 0 values are illegal for a position ina file.S257 EOF in #include directive
End of file was encountered while processing a #include directive.S258 Unmatched #elif
A #elif directive was encountered with no preceding #if or #elif directive.S259 Unmatched #else
A #else directive was encountered with no preceding #if or #elif directive.S260 Unmatched #endif
A #endif directive was encountered with no preceding #if, #ifdef, or #ifndef directive.S261 Include files nested too deeply
The nesting depth of #include directives exceeded the maximum (currently 20).S262 Unterminated macro definition for $
A newline was encountered in the formal parameter list for the indicated macro.S263 Unterminated string or character constant
A newline with no preceding backslash was found in a quoted string.
Messages
PVF Reference Guide Version 2017 | 193
I264 Possible nested comment
The characters /* were found within a comment.S265 <reserved message number>
S266 <reserved message number>
S267 <reserved message number>
W268 Cannot inline subprogram; common block mismatch
W269 Cannot inline subprogram; argument type mismatch
This message may be severe if the compilation has gone too far to undo the inliningprocess.F270 Missing -exlib option
W271 Can't inline $ - wrong number of arguments
I272 Argument of inlined function not used
S273 Inline library not specified on command line (-inlib switch)
F274 Unable to access file $/TOC
S275 Unable to open file $ while extracting or inlining
F276 Assignment to constant actual parameter in inlined subprogram
I277 Inlining of function $ may result in recursion
S278 <reserved message number>
W279 Possible use of $ before definition in $
The optimizer has detected the possibility that a variable is used before it has beenassigned a value. The names of the variable and the function in which the use occurredare listed. The line number, if specified, is the line number of the basic block containingthe use of the variable.W280 Syntax error in directive $
Messages 280-300 reserved for directives handlingW281 Directive ignored - $ $
S300 Too few data constants in initialization of derived type $
S301 $ must be TEMPLATE or PROCESSOR
S302 Unmatched END$ statement
S303 END statement for $ required in an interface block
S304 EXIT/CYCLE statement must appear in a DO/DOWHILE loop$$
S305 $ cannot be named, $
Messages
PVF Reference Guide Version 2017 | 194
S306 $ names more than one construct
S307 $ must have the construct name $
S308 DO may not terminate at an EXIT, CYCLE, RETURN, STOP, GOTO, or arithmetic IF
S309 Incorrect name, $, specified in END statement
S310 $ $
Generic message for MODULE errors.W311 Non-replicated mapping for $ array, $, ignored
W312 Array $ should be declared SEQUENCE
W313 Subprogram $ called within INDEPENDENT loop not PURE
E314 IPA: actual argument $ is a label, but dummy argument $ is not an asterisk
The call passes a label to the subprogram; the corresponding dummy argument in thesubprogram should be an asterisk to declare this as the alternate return.I315 IPA: routine $, $ constant dummy arguments
This many dummy arguments are being replaced by constants due to interproceduralanalysis.I316 IPA: routine $, $ INTENT(IN) dummy arguments
This many dummy arguments are being marked as INTENT(IN) due to interproceduralanalysis.I317 IPA: routine $, $ array alignments propagated
This many array alignments were propagated by interprocedural analysis.I318 IPA: routine $, $ distribution formats propagated
This many array distribution formats were propagated by interprocedural analysis.I319 IPA: routine $, $ distribution targets propagated
This many array distribution targets were propagated by interprocedural analysis.I320 IPA: routine $, $ common blocks optimized
This many mapped common blocks were optimized by interprocedural analysis.I321 IPA: routine $, $ common blocks not optimized
This many mapped common blocks were not optimized by interprocedural analysis,either because they were declared differently in different routines, or they did notappear in the main program.I322 IPA: analyzing main program $
Interprocedural analysis is building the call graph and propagating information with thenamed main program.I323 IPA: collecting information for $
Interprocedural analysis is saving information for the current subprogram forsubsequent analysis and propagation.
Messages
PVF Reference Guide Version 2017 | 195
W324 IPA file $ appears to be out of date
W325 IPA file $ is for wrong subprogram: $
W326 Unable to open file $ to propagate IPA information to $
I327 IPA: $ subprograms analyzed
I328 IPA: $ dummy arguments replaced by constants
I329 IPA: $ INTENT(IN) dummy arguments should be INTENT(INOUT)
I330 IPA: $ dummy arguments changed to INTENT(IN)
I331 IPA: $ inherited array alignments replaced
I332 IPA: $ transcriptive distribution formats replaced
I333 IPA: $ transcriptive distribution targets replaced
I334 IPA: $ descriptive/prescriptive array alignments verified
I335 IPA: $ descriptive/prescriptive distribution formats verified
I336 IPA: $ descriptive/prescriptive distribution targets verified
I337 IPA: $ common blocks optimized
I338 IPA: $ common blocks not optimized
S339 Bad IPA contents file: $
S340 Bad IPA file format: $
S341 Unable to create file $ while analyzing IPA information
S342 Unable to open file $ while analyzing IPA information
S343 Unable to open IPA contents file $
S344 Unable to create file $ while collecting IPA information
F345 Internal error in $: table overflow
Analysis failed due to a table overflowing its maximum size.W346 Subprogram $ appears twice
The subprogram appears twice in the same source file; IPA will ignore the firstappearance.F347 Missing -ipalib option
Interprocedural analysis, enabled with the -ipacollect, -ipaanalyze, or -ipapropagateoptions, requires the -ipalib option to specify the library directory.W348 Common /$/ $ has different distribution target
Messages
PVF Reference Guide Version 2017 | 196
The array was declared in a common block with a different distribution target in anothersubprogram.W349 Common /$/ $ has different distribution format
The array was declared in a common block with a different distribution format inanother subprogram.W350 Common /$/ $ has different alignment
The array was declared in a common block with a different alignment in anothersubprogram.W351 Wrong number of arguments passed to $
The subroutine or function statement for the given subprogram has a different numberof dummy arguments than appear in the call.W352 Wrong number of arguments passed to $ when bound to $
The subroutine or function statement for the given subprogram has a different numberof dummy arguments than appear in the call to the EXTERNAL name given.W353 Subprogram $ is missing
A call to a subroutine or function with this name appears, but it could not be found oranalyzed.I354 Subprogram $ is not called
No calls to the given subroutine or function appear anywhere in the program.W355 Missing argument in call to $
A nonoptional argument is missing in a call to the given subprogram.I356 Array section analysis incomplete
Interprocedural analysis for array section arguments is incomplete; some informationmay not be available for optimization.I357 Expression analysis incomplete
Interprocedural analysis for expression arguments is incomplete; some information maynot be available for optimization.W358 Dummy argument $ is EXTERNAL, but actual is not subprogram
The call statement passes a scalar or array to a dummy argument that is declaredEXTERNAL.W359 SUBROUTINE $ passed to FUNCTION dummy argument $
The call statement passes a subroutine name to a dummy argument that is used as afunction.W360 FUNCTION $ passed to FUNCTION dummy argument $ with different resulttype
The call statement passes a function argument to a function dummy argument, but thedummy has a different result type.W361 FUNCTION $ passed to SUBROUTINE dummy argument $
Messages
PVF Reference Guide Version 2017 | 197
The call statement passes a function name to a dummy argument that is used as asubroutine.W362 Argument $ has a different type than dummy argument $
The type of the actual argument is different than the type of the corresponding dummyargument.W363 Dummy argument $ is a POINTER but actual argument $ is not
The dummy argument is a pointer, so the actual argument must be also.W364 Array or array expression passed to scalar dummy argument $
The actual argument is an array, but the dummy argument is a scalar variable.W365 Scalar or scalar expression passed to array dummy argument $
The actual argument is a scalar variable, but the dummy argument is an array.F366 Internal error: interprocedural analysis fails
An internal error occurred during interprocedural analysis; please report this tothe compiler maintenance group. If user errors were reported when collecting IPAinformation or during IPA analysis, correcting them may avoid this error.I367 Array $ bounds cannot be matched to formal argument
Passing a nonsequential array to a sequential dummy argument may require copyingthe array to sequential storage. The most common cause is passing an ALLOCATABLEarray or array expression to a dummy argument that is declared with explicit bounds.Declaring the dummy argument as assumed shape, with bounds (:,:,:), will remove thiswarning.W368 Array-valued expression passed to scalar dummy argument $
The actual argument is an array-valued expression, but the dummy argument is a scalarvariable.W369 Dummy argument $ has different rank than actual argument
The actual argument is an array or array-valued expression with a different rank thanthe dummy argument.W370 Dummy argument $ has different shape than actual argument
The actual argument is an array or array-valued expression with a different shape thanthe dummy argument; this may require copying the actual argument into sequentialstorage.W371 Dummy argument $ is INTENT(IN) but may be modified
The dummy argument was declared as INTENT(IN), but analysis has found that theargument may be modified; the INTENT(IN) declaration should be changed.W372 Cannot propagate alignment from $ to $
The most common cause is when passing an array with an inherited alignment to adummy argument with non- inherited alignment.I373 Cannot propagate distribution format from $ to $
The most common cause is when passing an array with a transcriptive distributionformat to a dummy argument with prescriptive or descriptive distribution format.
Messages
PVF Reference Guide Version 2017 | 198
I374 Cannot propagate distribution target from $ to $
The most common cause is when passing an array with a transcriptive distributiontarget to a dummy argument with prescriptive or descriptive distribution target.I375 Distribution format mismatch between $ and $
Usually this arises when the actual and dummy arguments are distributed in differentdimensions.I376 Alignment stride mismatch between $ and $
This may arise when the actual argument has a different stride in its alignment to itstemplate than does the dummy argument.I377 Alignment offset mismatch between $ and $
This may arise when the actual argument has a different offset in its alignment to itstemplate than does the dummy argument.I378 Distribution target mismatch between $ and $
This may arise when the actual and dummy arguments have different distribution targetsizes.I379 Alignment of $ is too complex
The alignment specification of the array is too complex for interprocedural analysis toverify or propagate; the program will work correctly, but without the benefit of IPA.I380 Distribution format of $ is too complex
The distribution format specification of the array is too complex for interproceduralanalysis to verify or propagate; the program will work correctly, but without the benefitof IPA.I381 Distribution target of $ is too complex
The distribution target specification of the array is too complex for interproceduralanalysis to verify or propagate; the program will work correctly, but without the benefitof IPA.I382 IPA: $ subprograms analyzed
Interprocedural analysis succeeded in finding and analyzing this many subprograms inthe whole program.I383 IPA: $ dummy arguments replaced by constants
Interprocedural analysis has found this many dummy arguments in the whole programthat can be replaced by constants.I384 IPA: $ dummy arguments changed to INTENT(IN)
Interprocedural analysis has found this many dummy arguments in the whole programthat are not modified and can be declared as INTENT(IN).W385 IPA: $ INTENT(IN) dummy arguments should be INTENT(INOUT)
Interprocedural analysis has found this many dummy arguments in the whole programthat were declared as INTENT(IN) but should be INTENT(INOUT).I386 IPA: $ array alignments propagated
Messages
PVF Reference Guide Version 2017 | 199
Interprocedural analysis has found this many array dummy arguments that could havethe inherited array alignment replaced by a descriptive alignment.I387 IPA: $ array alignments verified
Interprocedural analysis has verified that the prescriptive or descriptive alignments ofthis many array dummy arguments match the alignments of the actual argument.I388 IPA: $ array distribution formats propagated
Interprocedural analysis has found this many array dummy arguments that could havethe transcriptive distribution format replaced by a descriptive format.I389 IPA: $ array distribution formats verified
Interprocedural analysis has verified that the prescriptive or descriptive distributionformats of this many array dummy arguments match the formats of the actualargument.I390 IPA: $ array distribution targets propagated
Interprocedural analysis has found this many array dummy arguments that could havethe transcriptive distribution target replaced by a descriptive target.I391 IPA: $ array distribution targets verified
Interprocedural analysis has verified that the prescriptive or descriptive distributiontargets of this many array dummy arguments match the targets of the actual argument.I392 IPA: $ common blocks optimized
Interprocedural analysis has found this many common blocks that could be optimized.I393 IPA: $ common blocks not optimized
Interprocedural analysis has found this many common blocks that could not beoptimized, either because the common block was not declared in the main program, orbecause it was declared differently in different subprograms.I394 IPA: $ replaced by constant value
The dummy argument was replaced by a constant as per interprocedural analysis.I395 IPA: $ changed to INTENT(IN)
The dummy argument was changed to INTENT(IN) as per interprocedural analysis.I396 IPA: array alignment propagated to $
The template alignment for the dummy argument was changed as per interproceduralanalysis.I397 IPA: distribution format propagated to $
The distribution format for the dummy argument was changed as per interproceduralanalysis.I398 IPA: distribution target propagated to $
The distribution target for the dummy argument was changed as per interproceduralanalysis.I399 IPA: common block $ not optimized
Messages
PVF Reference Guide Version 2017 | 200
The given common block was not optimized by interprocedural analysis either becauseit was not declared in the main program, or because it was declared differently indifferent subprograms.E400 IPA: dummy argument $ is an asterisk, but actual argument is not a label
The subprogram expects an alternate return label for this argument.E401 Actual argument $ is a subprogram, but Dummy argument $ is not declaredEXTERNAL
The call statement passes a function or subroutine name to a dummy argument that is ascalar variable or array.E402 Actual argument $ is illegal
E403 Actual argument $ and formal argument $ have different ranks
The actual and formal array arguments differ in rank, which is allowed only if botharrays are declared with the HPF SEQUENCE attribute.E404 Sequential array section of $ in argument $ is not contiguous
When passing an array section to a formal argument that has the HPF SEQUENCEattribute, the actual argument must be a whole array with the HPF SEQUENCEattribute, or an array section of such an array where the section is a contiguous sequenceof elements.E405 Array expression argument $ may not be passed to sequential dummy argument $
When the dummy argument has the HPF SEQUENCE attribute, the actual argumentmust be a whole array with the HPF SEQUENCE attribute or a contiguous array sectionof such an array, unless an INTERFACE block is used.E406 Actual argument $ and formal argument $ have different character lengths
The actual and formal array character arguments have different character lengths, whichis allowed only if both character arrays are declared with the HPF SEQUENCE attribute,unless an INTERFACE block is used.W407 Argument $ has a different character length than dummy argument $
The character length of the actual argument is different than the length specified for thecorresponding dummy argument.W408 Specified main program $ is not a PROGRAM
The main program specified on the command line is a subroutine, function, or blockdata subprogram.W409 More than one main program in IPA directory: $ and $
There is more than one main program analyzed in the IPA directory shown. The first onefound is used.W410 No main program found; IPA analysis fails.
The main program must appear in the IPA directory for analysis to proceed.W411 Formal argument $ is DYNAMIC but actual argument is an expression
W412 Formal argument $ is DYNAMIC but actual argument $ is not
Messages
PVF Reference Guide Version 2017 | 201
I413 Formal argument $ has two reaching distributions and may be a candidate forcloning
I414 $ and $ may be aliased and one of them is assigned
Interprocedural analysis has determined that two formal arguments may be aliasedbecause the same variable is passed in both argument positions; or one formal argumentand a global or COMMON variable may be aliased, because the global or COMMONvariable is passed as an actual argument. If either alias is assigned in the subroutine,unexpected results may occur; this message alerts the user that this situation isdisallowed by the Fortran standard.F415 IPA fails: incorrect IPA file
Interprocedural analysis saves its information in special IPA files in the specified IPAdirectory. One of these files has been renamed or corrupted. This can arise when thereare two files with the same prefix, such as a.hpf and a.f90.E416 Argument $ has the SEQUENCE attribute, but the dummy parameter $ does not
When an actual argument is an array with the SEQUENCE attribute, the dummyparameter must have the SEQUENCE attribute or an INTERFACE block must be used.E417 Interface block for $ is a SUBROUTINE but should be a FUNCTION
E418 Interface block for $ is a FUNCTION but should be a SUBROUTINE
E419 Interface block for $ is a FUNCTION has wrong result type
W420 Earlier $ directive overrides $ directive
W421 $ directive can only appear in a function or subroutine
E422 Nonconstant DIM= argument is not supported
E423 Constant DIM= argument is out of range
E424 Equivalence using substring or vector triplets is not allowed
E425 A record is not allowed in this context
E426 WORD type cannot be converted
E427 Interface block for $ has wrong number of arguments
E428 Interface block for $ should have $
E429 Interface block for $ should not have $
E430 Interface block for $ has wrong $
W431 Program is too large for Interprocedural Analysis to complete
W432 Illegal type conversion $
E433 Subprogram $ called within INDEPENDENT loop not LOCAL
Messages
PVF Reference Guide Version 2017 | 202
W434 Incorrect home array specification ignored
W435 Array declared with zero size
An array was declared with a zero or negative dimension bound, as ‘real a(-1)’, or anupper bound less than the lower bound, as ‘real a(4:2)’.W436 Independent loop not parallelized$
W437 Type $ will be mapped to $
Where DOUBLE PRECISION is not supported, it is mapped to REAL, and similarly forCOMPLEX(16) or COMPLEX*32.E438 $ $ not supported on this platform
This construct is not supported by the compiler for this target.S439 An internal subprogram cannot be passed as argument - $
S440 Defined assignment statements may not appear in WHERE statement or WHEREblock
S441 $ may not appear in a FORALL block
E442 Adjustable-length character type not supported on this host - $ $
S443 EQUIVALENCE of derived types not supported on this host - $
S444 Derived type in EQUIVALENCE statement must have SEQUENCE attribute - $
A variable or array with derived type appears in an EQUIVALENCE statement. Thederived type must have the SEQUENCE attribute, but does not.E445 Array bounds must be integer $ $
The expressions in the array bounds must be integer.S446 Argument number $ to $: rank mismatch
The number of dimensions in the array or array expression does not match the numberof dimensions in the dummy argument.S447 Argument number $ to $ must be a subroutine or function name
S448 Argument number $ to $ must be a subroutine name
S449 Argument number $ to $ must be a function name
S450 Argument number $ to $: kind mismatch
S451 Arrays of derived type with a distributed member are not supported
S452 Assumed length character, $, is not a dummy argument
S453 Derived type variable with pointer member not allowed in IO - $ $
S454 Subprogram $ is not a module procedure
Only names of module procedures declared in this module or accessed through USEassociation can appear in a MODULE PROCEDURE statement.
Messages
PVF Reference Guide Version 2017 | 203
S455 A derived type array section cannot appear with a member array section - $
A reference like A(:)%B(:), where ‘A’ is a derived type array and ‘B’ is a member array, isnot allowed; a section subscript may appear after ‘A’ or after ‘B’, but not both.S456 Unimplemented for data type for MATMUL
S457 Illegal expression in initialization
S458 Argument to NULL() must be a pointer
S459 Target of NULL() assignment must be a pointer
S460 ELEMENTAL procedures cannot be RECURSIVE
S461 Dummy arguments of ELEMENTAL procedures must be scalar
S462 Arguments and return values of ELEMENTAL procedures cannot have thePOINTER attribute
S463 Arguments of ELEMENTAL procedures cannot be procedures
S464 An ELEMENTAL procedure cannot be passed as argument - $
S465 Functions returning a POINTER require an explicit interface
S466 Member $ of derived type $ has PRIVATE type
S467 Target of NULL() assignment must have the ALLOCATABLE attribute
W468 Argument to ISO_C_BINDING intrinsic must have TARGET attribute set
W469 Character argument to C_LOC intrinsic must have length of one
W470 Accelerator feature license not found; accelerator features disabled
W471 CUDA Fortran feature license not found; CUDA Fortran features disabled
E472 A Scalar element of a nonsequential array cannot be passed to a dummy arrayargument - $
A subroutine or function call may not pass an element of an array, like 'A(N)', to adummy array argument if the array 'A' is not sequential. If the array is sequential, thenFortran sequence and storage association rules will treat the dummy argument as a newarray equivalenced to the actual argument starting at the element passed. If the array isnot sequential, then Fortran sequence and storage association rules do not apply.W473 $ must have the PURE attribute
F474 This type EXTRINSIC is not yet implemented - $
Contact PGI to ask when this EXTRINSIC type will be implemented.E475 A dummy argument may not be distributed in a PURE interface - $
A dummy argument to a routine defined with a PURE interface may not have theDISTRIBUTE attribute.
Messages
PVF Reference Guide Version 2017 | 204
E476 A dummy argument may only be aligned with another dummy in a PUREinterface - $
E477 The device array section actual argument was not stride-1 in the leading dimension- $
A device (device, shared, or constant attribute) array passed as an array section to anassumed-shape dummy argument must be stride-1 in the leading dimension.E478 Invalid actual argument to REFLECTED dummy argument - $
The actual argument symbol or expression to a dummy argument with the AcceleratorREFLECTED attribute must be a symbol that has a visible device copy. Expressions arenot allowed.E479 The dummy argument $ is REFLECTED; the actual argument $ must have a visibledevice copy
If a dummy argument has the Accelerator REFLECTED attribute, the actual argumentmust be a symbol with a visible device copy. This may be because the symbol appearedin a MIRROR, REFLECTED, COPYIN, COPYOUT, COPY or LOCAL declarativeAccelerator directive, or because it appeared in a COPYIN, COPYOUT, COPY or LOCALclause for an Accelerator DATA REGION or REGION surrounding the procedure call.E480 Argument $ is passed to dummy argument $, which is REFLECTED; the actualargument must not require runtime reshaping
When an actual argument is an array section or pointer array section, sometimes theactual argument must be copied to a temporary array. This may occur if the dummyargument is not assumed-shape, and so must be contiguous in memory, or if the actualargument is not stride-1 in the leftmost (first) dimension. In these cases, the REFLECTEDargument is not supported.F481 An ENTRY name must not appear as a dummy argument - $
The name of the subprogram or an ENTRY to the subprogram must not appear as adummy argument to the subprogram.482 COMMON /$/ is declared differently in two subprograms - $
The COMMON block name was declared with different distribution or alignment forone or more members in two different subprograms.E483 Storage association due to EQUIVALENCE($,$) causes HPF alignments anddistributions to be ignored
An EQUIVALENCE statement causes Fortran storage association between entriesin this COMMON block. The storage association overrides the HPF alignments anddistributions for the COMMON block members.E484 Datatype conflict in EQUIVALENCE between two distributed or alignedCOMMON block members: $ and $
Two distributed COMMON block members that appear in a COMMON block must havethe same datatype.E485 Datatype conflict in EQUIVALENCE between a distributed or aligned COMMONblock member and another: $ and $
Messages
PVF Reference Guide Version 2017 | 205
A distributed COMMON block member may not be EQUIVALENCEd with anotherCOMMON member.E486 The dummy argument $ is REFLECTED; an array element cannot be passed to aREFLECTED argument
An actual argument that is an array element cannot be passed to a REFLECTED dummyargument.E487 Index variable $ does not appear in a subscript on the left hand side of the FORALLassignment
In a FORALL statement, each index variable in the FORALL must appear in somesubscript of the left hand side of the FORALL assignment. Otherwise, the FORALL willassign the same left hand side elements for different values of that index.I489 An ALLOCATE of a POINTER with transcriptive or inherited distribution causesreplication - $
When an array with the POINTER attribute and with a distribution that is transcriptiveor inherited is allocated, the alignment and distribution are ignored and the arraypointer is treated as replicated, since there is no symbol from which to inherit adistribution.E488 The function call in the FORALL does not have the PURE attribute - $
In a FORALL statement, all functions used must be PURE or ELEMENTAL. Otherwise,they cannot be called in parallel.E490 An array section of $ is passed to the REFLECTED argument $, which is notsupported
When an actual argument is an array section, the dummy argument must not have theREFLECTED attribute.W491 EXTRINSIC($) subprograms require an explicit interface - $
An EXTRINSIC subprogram with the LOCAL or SERIAL attributes require an explicitinterface, either through an INTERFACE block, or by being in the same MODULE as thecaller, or being in a MODULE that is referenced with a USE statement.E492 DYNAMIC distribution is only supported in HPF_GLOBAL subprograms - $
Variables with DYNAMIC distribution are not supported in EXTRINSIC(F77_LOCAL),EXTRINSIC(F77_SERIAL), EXTRINSIC(F90_LOCAL), EXTRINSIC(F90_SERIAL),EXTRINSIC(HPF_LOCAL) or EXTRINSIC(HPF_SERIAL) subprograms.E493 $ arrays may not be aligned with ALLOCATABLE arrays - $
Static local arrays, common arrays, and dummy argument arrays may not be alignedwith arrays that have the ALLOCATABLE attribute, since the allocatable alignee maynot be allocated.E494 COMMON arrays may not be aligned with dummy argument arrays - $
An array in a COMMON block may not specify an alignment with a dummy argumentarray.W495 The SHADOW directive for CYCLIC distributed dimensions is ignored - $
Messages
PVF Reference Guide Version 2017 | 206
A shadow boundary specified for array dimensions that are distributed with theCYCLIC distribution is ignored.I496 A $ of an unused template is eliminated
The HPF executable REDISTRIBUTE or REALIGN directive appeared specifying an HPFTEMPLATE that is not used; the REDISTRIBUTE or REALIGN is eliminated.E497 EXTRINSIC(F77_LOCAL) does not support distributed symbols of this datatype - $
This HPF implementation does not support distributed symbols of character or derivedtype in EXTRINSIC(F77_LOCAL) subprograms.E498 Alignment cycle involving two or more arguments - $
This dummy argument appears in an HPF ALIGN directive specifying alignment toanother dummy argument that is then aligned to this argument, or aligned to anotherdummy argument that is eventually aligned to this argument.W499 The descriptive distribution or alignment for this dummy argument is treated asprescriptive - $
Even though the distribution or alignment for this dummy argument was specified asdescriptive, it is treated as prescriptive.E500 MODULE $ uses (directly or indirectly) MODULE $, which causes a USE cycle
If MODULE A has a USE statement for MODULE B, we say that MODULE A directlyuses MODULE B. If MODULE B has a USE statement for MODULE C, we say thatMODULE A indirectly uses MODULE C. If MODULE C then has a USE statement forMODULE A, then MODULE A indirectly uses itself, which is a USE cycle, and is notallowed.E504 DIM argument out of range for this symbol - $
The DIM argument to this transformation intrinsic (CSHIFT, EOSHIFT, ...) must bebetween 1 and the rank of the array or expression being transformed.E505 DIM argument out of range for this reduction - $
The DIM argument to this reduction intrinsic (SUM, PRODUCT, ...) must be between 1and the rank of the expression being reduced.E506 The argument to ASSOCIATED must be a pointer - $
The argument to the ASSOCIATED intrinsic function must be a variable or array withthe POINTER attribute.E507 The arguments to MOVE_ALLOC must be ALLOCATABLE - $
The arguments to the MOVE_ALLOC procedure must have the ALLOCATABLEattribute.E508 The array objects in a call to an elemental function are not conformable - $
When calling an elemental function, the arguments must be scalars or conformablearrays or array expressions.E509 Variables in a PURE subprogram may not have the SAVE attribute - $
PURE subprograms cannot refer to external, module, or COMMON data, and cannotsave state in a SAVEd variable.
Messages
PVF Reference Guide Version 2017 | 207
E510 Only assignment statements are allowed in a WHERE construct
A WHERE construct is the WHERE statement and all the statements until the matchingENDWHERE. The body of the WHERE construct can only contain assignmentstatements.E511 The WHERE mask expression and the array assignment do not conform
The assignment under control of a WHERE mask must have the same shape as theWHERE mask.E512 The WHERE mask is not an array expression
The WHERE mask expression must be a logical array expression.E513 The alignment or distribution target may not be a private variable - $
This is a HPF_CRAFT restriction.E514 The alignment extends beyond the bounds of the template - $
When aligning to a template, the entire array must align to template elements that liewithin the bounds of the template.E515 Static variable aligned with allocatable symbol - $
A nonallocatable symbol cannot be aligned to an allocatable symbol.E516 PURE subprograms may not have distributed variables - $
Distributed arrays are not allowed in PURE subprograms.E517 Variables in HPF_LOCAL subprograms may not be distributed - $
Distributed arrays are not allowed in HPF_LOCAL subprograms.W518 Function result could not be distributed; replicating - $
The compiler will replicate the function result.E519 More than one device-resident object in assignment
Only one device-resident variable or array is allowed in an assignment.E520 Host MODULE data cannot be used in a DEVICE or GLOBAL subprogram - $
CUDA Fortran DEVICE or GLOBAL subprograms cannot access host data directly.E521 MODULE data cannot be used in a DEVICE or GLOBAL subprogram unlesscompiling for compute capability >= 2.0 - $
CUDA Fortran DEVICE or GLOBAL subprograms cannot access data from anyMODULE except the MODULE containing the subprogram, unless they are beingcompiled for compute capability 2.0 or higher. This feature requires the unified memorysystem provided in compute capability 2.0.E522 MODULE data cannot be used in a DEVICE or GLOBAL subprogram unlesscompiling with CUDA Toolkit 3.0 or later - $
CUDA Fortran DEVICE or GLOBAL subprograms cannot access data from anyMODULE except the MODULE containing the subprogram, unless they are beingcompiled for compute capability 2.0 or higher with the CUDA Toolkit 3.0 or later.
This feature requires the unified memory system provided in compute capability 2.0.
Messages
PVF Reference Guide Version 2017 | 208
W523 MODULE data used in a DEVICE or GLOBAL subprogram forces computecapability >= 2.0 only - $
CUDA Fortran DEVICE or GLOBAL subprograms can access MODULE data only whencompiled for compute capability 2.0 or greater.E524 Dependency in assignment causes allocation of a temporary which is notsupported in DEVICE or GLOBAL subprograms
The compiler has identified a possible dependency in an assignment statement whichrequires allocation of temporary storage to produce a correct result. Dynamic allocationof memory is not supported in subprograms that run on the device.E525 Array reshaping is not supported for device subprogram calls: argument $ tosubprogram $
Passing an array section or assumed-shape array to a non-assumed-shape dummyargument is not supported in global or device subprograms. This would require a run-time test and a possible run-time copy to a dynamically allocated temporary array.W526 SHARED attribute ignored on dummy argument $
The SHARED attribute has no meaning when applied to a dummy argument.E527 Argument number $ requires allocation of a temporary which is not supported inDEVICE or GLOBAL subprograms
Evaluation of the specified argument requires allocation of temporary storage for theresult to be passed to the subprogram being called. Dynamic allocation of memory is notsupported in subprograms that run on the device.E528 Argument number $ to $: device attribute mismatch
Device attributes of the actual and formal arguments are not the same.E529 PRINT and WRITE statements in device subprograms are only supported whencompiling with CUDA Toolkit 4.0 or later
Support for PRINT * or WRITE(*,*) statements in CUDA Fortran device subprogramsrequires CUDA Toolkit 4.0 or later and compute capability 2.0 or higher.E530 PRINT and WRITE statements in device subprograms are only supported withcompute capability 2.0 or higher
Support for PRINT * or WRITE(*,*) statements in CUDA Fortran device subprogramsrequires CUDA Toolkit 4.0 or later and compute capability 2.0 or higher.W531 PGI extension to OpenACC: $
This program is using a PGI extension to OpenACC.W532 OpenACC feature not yet implemented: $
This OpenACC feature is not yet implemented. This program is using a PGI extension toOpenACC.E533 Clause $ not allowed in $ directive
This clause is not allowed on the specified directive.E534 A loop scheduling directive may not appear within a KERNEL loop
Messages
PVF Reference Guide Version 2017 | 209
An accelerator or OpenACC loop directive that specifies a schedule, such as PARALLEL,VECTOR, WORKER or GANG, may not appear inside a loop that has an acceleratorloop directive with the KERNEL clause. This clause is not allowed on the specifieddirective.E535 Undeclared symbol $ used in directive
Symbols used in OpenACC directives must be declared.S901 #elif after #else
A preprocessor #elif directive was found after a #else directive; only #endif is allowed inthis context.S902 #else after #else
A preprocessor #else directive was found after a #else directive; only #endif is allowed inthis context.W905 Argument mismatch for $
The number of arguments supplied in the call to the indicated macro did not agree withthe number of parameters in the macro's definition.F906 Can't find include file $
The indicated include file could not be opened.S908 EOFin comment
The end of a file was encountered while processing a comment.S909 EOFin macro call to $
The end of a file was encountered while processing a call to the indicated macro.S912 Identifier too long
The length of an identifier exceeded the maximum allowed (currently 2048).W914 Illegal directive name
The sequence of characters following a # sign was not an identifier.W915 Illegal macro name
A macro name was not an identifier.W918 Missing #endif
End of file was encountered before a required #endif directive was found.W919 Missing argument list for $
A call of the indicated macro had no argument list.S920 Number too long
The length of a number exceeded the maximum allowed (currently 2048).W921 Redefinition of symbol $
The indicated macro name was redefined.I922 Redundant definition for symbol $
A definition for the indicated macro name was found that was the same as a previousdefinition.
Messages
PVF Reference Guide Version 2017 | 210
F923 String too long
The length of a quoted string exceeded the maximum allowed (currently 2048).S924 Syntax error in #define, formal $ not identifier
A formal parameter that was not an identifier was used in a macro definition.S926 Syntax error in #if
A syntax error was found while parsing the expression following a #if or #elif directive.S927 Syntax error in #include
The #include directive was not correctly formed.W928 Syntax error in #line
A #line directive was not correctly formed.W929 Syntax error in #module
A #module directive was not correctly formed.W930 Syntax error in #undef
A #undef directive was not correctly formed.W931 Token after #ifdef must be identifier
The #ifdef directive was not followed by an identifier.W932 Token after #ifndef must be identifier
The #ifndef directive was not followed by an identifier.S933 Too many actual parameters to $
The number of actual arguments to the indicated macro exceeded the maximum allowed(currently 31).S934 Too many formal parameters to $
The number of formal arguments to the indicated macro exceeded the maximumallowed (currently 31).S935 Illegal context for __VA_ARGS__
W936 Undefined directive $
The identifier following a # was not a directive name.S937 EOFin #include directive
End of file was encountered while processing a #include directive.S938 Unmatched #elif
A #elif directive was encountered with no preceding #if or #elif directive.S939 Unmatched #else
A #else directive was encountered with no preceding #if or #elif directive.S940 Unmatched #endif
A #endif directive was encountered with no preceding #if, #ifdef, or #ifndef directive.W941 Illegal token in directive, $
Messages
PVF Reference Guide Version 2017 | 211
A directive token contains a illegal character.S942 Unterminated macro definition for $
A newline was encountered in the formal parameter list for the indicated macro.S943 Unterminated string or character constant
A newline with no preceding backslash was found in a quoted string.I944 Possible nested comment
The characters /* were found within a comment.I945 Redefining predefined macro $
I946 Undefining predefined macro $
W947 Can't redefine predefined macro $
W948 Can't undefine predefined macro $
F949 #error -- $
User defined preprocessor error message.W950 #ident not followed by quoted string
W951 Extraneous tokens ignored following # directive
F952 Unexpected EOF following #directive
W953 Unexpected # ignored in #if expression
S954 Illegal number in directive
S955 Illegal token in #if expression
S956 Missing > in #include
W957 Arguments in macro $ are not unique
S959 ## directive occurs at beginning or end of macro definition
S960 $ is not an argument
W961 No macro replacement within a character constant
W962 Macro replacement within a character constant
W964 Macro replacement within a string literal
F965 Recursive include file $
W966 Null argument to macro
Argument to macro is a null value.F967 #warning -- $
User defined preprocessor warning message.
Messages
PVF Reference Guide Version 2017 | 212
S969 _Pragma $
Pragma operator errors.
8.4. Fortran Run-time Error MessagesThis section presents the error messages generated by the run-time system. The run-timesystem displays error messages on standard output.
8.4.1. Message FormatThe messages are numbered but have no severity indicators because they all terminateprogram execution.
8.4.2. Message ListHere are the run-time error messages:
201 illegal value for specifier
An improper specifier value has been passed to an I/O run-time routine. Example:within an OPEN statement, form='unknown'.
202 conflicting specifiers
Conflicting specifiers have been passed to an I/O run-time routine. Example: within anOPEN statement, form='unformatted', blank='null'.
203 record length must be specified
A recl specifier required for an I/O run-time routine has not been passed. Example:within an OPEN statement, access='direct' has been passed, but the record length has notbeen specified (recl=specifier).
204 illegal use of a readonly file
Self explanatory. Check file and directory modes for readonly status.
205 'SCRATCH' and 'SAVE'/'KEEP' both specified
In an OPEN statement, a file disposition conflict has occurred. Example: within anOPEN statement, status='scratch' and dispose='keep' have both been passed.
206 attempt to open a named file as 'SCRATCH'
207 file is already connected to another unit
208 'NEW' specified for file that already exists
209 'OLD' specified for file that does not exist
210 dynamic memory allocation failed
Messages
PVF Reference Guide Version 2017 | 213
Memory allocation operations occur only in conjunction with namelist I/O. Themost probable cause of fixed buffer overflow is exceeding the maximum number ofsimultaneously open file units.
211 invalid file name
212 invalid unit number
A file unit number less than or equal to zero has been specified.
215 formatted/unformatted file conflict
Formatted/unformatted file operation conflict.
217 attempt to read past end of file
219 attempt to read/write past end of record
For direct access, the record to be read/written exceeds the specified record length.
220 write after last internal record
221 syntax error in format string
A run-time encoded format contains a lexical or syntax error.
222 unbalanced parentheses in format string
223 illegal P or T edit descriptor - value missing
224 illegal Hollerith or character string in format
An unknown token type has been found in a format encoded at run-time.
225 lexical error -- unknown token type
226 unrecognized edit descriptor letter in format
An unexpected Fortran edit descriptor (FED) was found in a run-time format item.
228 end of file reached without finding group
229 end of file reached while processing group
230 scale factor out of range -128 to 127
Fortran P edit descriptor scale factor not within range of -128 to 127.
231 error on data conversion
233 too many constants to initialize group item
234 invalid edit descriptor
An invalid edit descriptor has been found in a format statement.
235 edit descriptor does not match item type
Messages
PVF Reference Guide Version 2017 | 214
Data types specified by I/O list item and corresponding edit descriptor conflict.
236 formatted record longer than 2000 characters
237 quad precision type unsupported
238 tab value out of range
A tab value of less than one has been specified.
239 entity name is not member of group
240 no initial left parenthesis in format string
241 unexpected end of format string
242 illegal operation on direct access file
243 format parentheses nesting depth too great
244 syntax error - entity name expected
245 syntax error within group definition
246 infinite format scan for edit descriptor
248 illegal subscript or substring specification
249 error in format - illegal E, F, G or D descriptor
250 error in format - number missing after '.', '-', or '+'
251 illegal character in format string
252 operation attempted after end of file
253 attempt to read non-existent record (direct access)
254 illegal repeat count in format
255 illegal asynchronous I/O operation
256 POS can only be specified for a 'STREAM' file
257 POS value must be positive
258 NEWUNIT requires FILE or STATUS=SCRATCH
PVF Reference Guide Version 2017 | 215
Chapter 9.CONTACT INFORMATION
You can contact PGI at:
20400 NW Amberwood Drive Suite 100Beaverton, OR 97006
Or electronically using any of the following means:
Fax: +1-503-682-2637Sales: mailto: [email protected]: https://www.pgicompilers.com
The PGI User Forum, https://www.pgicompilers.com/userforum/index.php is monitoredby members of the PGI engineering and support teams as well as other PGI customers.The forums contain answers to many commonly asked questions. Log in to the PGIwebsite, https://www.pgicompilers.com/account/login.php" to access the forums.
Many questions and problems can be resolved by following instructions andthe information available in the PGI frequently asked questions (FAQ), https://www.pgicompilers.com/support/faq.htm.
Submit support requests using the PGI Technical Support Request form, https://www.pgicompilers.com/support/support_request.php .
Notice
ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS,DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY,"MATERIALS") ARE BEING PROVIDED "AS IS." NVIDIA MAKES NO WARRANTIES,EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THEMATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OFNONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULARPURPOSE.
Information furnished is believed to be accurate and reliable. However, NVIDIACorporation assumes no responsibility for the consequences of use of suchinformation or for any infringement of patents or other rights of third partiesthat may result from its use. No license is granted by implication of otherwiseunder any patent rights of NVIDIA Corporation. Specifications mentioned in thispublication are subject to change without notice. This publication supersedes andreplaces all other information previously supplied. NVIDIA Corporation productsare not authorized as critical components in life support devices or systemswithout express written approval of NVIDIA Corporation.
Trademarks
NVIDIA, the NVIDIA logo, Cluster Development Kit, PGC++, PGCC, PGDBG, PGF77,PGF90, PGF95, PGFORTRAN, PGHPF, PGI, PGI Accelerator, PGI CDK, PGI Server,PGI Unified Binary, PGI Visual Fortran, PGI Workstation, PGPROF, PGROUP, PVF,and The Portland Group are trademarks and/or registered trademarks of NVIDIACorporation in the U.S. and other countries. Other company and product namesmay be trademarks of the respective companies with which they are associated.
Copyright
© 2013–2017 NVIDIA Corporation. All rights reserved.
PGI Compilers and Tools
Top Related